Title of Invention


Abstract ABSTRACT Various compounds obtained from plants of the Barringtonia species which are derived from Barringtoside A and Barringtoside C as precursor compounds which especially have an arabinopyranosyl substituent at the 21 position which may optionally be further substituted with benzoyl, dibenzoyl, methyl butanoyl, methyl butyryl or tigloyl at the 3 or 4 positions. Alternatively at the 21 position there is provided tigloyl, benzoyl or dibenzoyl substituents.
Full Text

This invention relates to novel compounds having analgesic properties and extracts containing same. Such compounds are obtained from plants of the Bamngtonia species.
BACKGROUND OF THE INVENTION Bamngtonia comprise the largest genus of plants within the family
Lecythfdaceae and are widely distributed Jn the tropical regions of Asia, Malaysia and the Pacific. [1]
Bamngtonia are trees or shrubs ranging in size from 2m to in excess of 25m. Four species are known to occur In Northern Australia [2], three of which, B. racemosa [(L.) Spreng], B. catypfrate [(Mietrs) R. Br. ex Bailey] and B. asiatica [(L.) Kurtz] are known only from north Queensland* The fourth Australian species, B. acutangula [(L.) Gaertn], has wider distribution, being found across Northern Australia from North Queensland to North Western Australia. Barringtonia acutangula has been further divided into two subspecies, B. acutanguia ssp. acutangula and B. aoutangufa ssp. sp/cafa [1]. The latter of these Is found throughout the Bamngtonia distribution area whereas the former is restricted to Northern Australia,
Throughout their range many Bamngtonia species have been used In variety of ways by local people. One common use of Barringtonfa species is as a fish poison [2-5], Several species have been reported as fish poisons Including B. acutangula [2-4, 6-11], B. spedosa [5], R racemosa [2, 3,10-12], B. asiatrica [2,3,5,10,11] and B. calyptrate [10]. Although the fruit and bark of the tree is often used as a fish poison [2,3,5-12], several other plant organs, including leaves [8,11], roots [4,8,9,11], seeds [2,9,10,12] and wood [12], have also been used. The leaves, fruit and seeds of several species are known to be edible [3,9,12-15],
Several other properties and uses of BairlngtonJa species have been reported. These include the use of B« racemosa as a tanning agent, due to

the presence of tannins, [3,12,16] and as an insecticide reportedly to be approximately half as potent as nicotine [12,16,17J. The fruftof Bamngtonia has been used to poison wild pig [12]. In addition the seed of 6, racemosa and the fruit of 0, asiatica has been used for suicide and administration with "... homicidal Intent..." [11,12], coconut milk being an antidote. These toxic properties may be due to the presence of HCN which has been demonstrated in high concentrations in the kernel of B. asiatica [11},
Many of the Baningtonia species have found extensive use as traditional medicines and the fruit of B. acutangula has been called "Nurse fruit? [6]. AH parts of the plant have been used and applications have been both internal and external. Preparation of applications may involve drying and powdering, extraction with hot or cold water, heating or juicing [3,11,12, 16,18], External applications tend to focus, as expected, on skin disease. Ailments such as general wounds, rheumatism, eczema, ulcers, scabies, tinea, ringworm, itches, inflammation and even leprosy have been treated with Bamngtonia species [3,11,12,16,18]. Seeds in powered form have been used as a snuff to relieve headache whereas heated seeds are aromatic and have been used to assist In colic and parturition [3], External applications to assist ophthalmia, chest cold and pain, asthma, fever, colic, flatulence, non-venereal stricture, sore throat and stomach ache have also been reported [3,6,11.12,16,18].
Common Internal used of Bamngtonia are for the relief of diarrhoea, dysentery and stomach ache, as an emetic, expectorate and laxative [3,6,8-12,16,18,19J. Preparations of some species are taken as a bitter tonic [3, 8,9,11,18,19] and the seed of B. rac&mosa is taken as a vermifuge [18]. As fish poisons, Baningtonia species, in particular B. actungula [A], were used extensively in Australia, however it seems that little use was made of Baningtonia species as medicines compared with other regions in which the plants are located. In Australian, Bamngtonia extracts have been used for skin complaints such as wounds, boils and chickenpox (B, racemosa and B. acutangula), for chest pain and fever (B, catyprata), In ophthalmia, colic, parturition and to Induce vomiting (R acutangula) [10, 11, 16, 18]. More

detailed accounts of the uses of Barrfngtonia sp. can be found in the literature (eg [3, 12,18]).
In view of the wide traditional application of Barringtonia species as medicinal plants, it is surprising that the chemical nature of the bioactive constituents has attracted little attention.
The presence of saponin-like glycosides in B. insingnis, B. vriesei and B. racemosa was demonstrated as early as 1898 (reported in [6]). Subsequently, In 1901, a saponln was isolated from B. speciosa which yielded, on hydrolysis, glucose and banfngtogenin (reported in [20], although [6] reports the species as B. spinosa). The same author also reported the presence of a second sapogenin, namely barringtogenttin. Nozoe Isolated Ai-barrinin and Ai-banigenta from the seeds of B. asfatica and subsequently reported that the saponin of Arbarrinin contained gluconic acid, o^glucose, d-galactose and a methyl pentose [21, 22], Alkafine hydrolysis of Ar barrigenin gave tigllc acid and a new aglycone, Ai-barrigenol [23]. Acetylation of Ai-banigenin also led to the Isolation of a second aglycone, A^bamgenol [23].
the presence of high concentrations of saponins was reported from the seeds, leaves and bark of B. acutangufa and B. racemosa and three sapogenins were identified 120}. Much of the ensuing work aimed to isolate and characterise the nature of these saponins.
The structure of Ai-barrigenol as first assigned by Cole et al. in 1955 [24] is shown in FIG 1.
During the 1950's, there was a growing interest in saponins and sapogenins as evidenced by the number of publications in which, using mainly degradative techniques, structures were assigned and revised. The first sapogenins isolated from a Banihgfon/a specfes after AT and Ay-barrfgenol were barringtogenol and barringtogenic acid which were isolated from the fruit of B. racemosa [25] and which structures are shown In FIG 2. Barringtonia acutangula continued to be a source of novel
saponins and sapogenins. Again from the fruit of this species, a series of

compounds, barringtogenoi B, C, D, and E, were isolated and their structures explored [8,26-34].
Barringtpgenol C was isolated from B. acutangula fruits and the structure assigned by chemical techniques as previously described (FIG 4) (eg [8f 26, 27. 31-34]).
Barringtogenoi D, again isolated from S. acutangula fruit, was described by Barua et al [26J and a structure proposed by Chakraborti and Barua [29,303 (FIG 5).
Barringtogenoi E was isolated from the branch wood of B. acutangula and a structure was assigned using mass spectral and chemical information (FIG 6) [8, 28J. It was noted that barringtogenol E was perhaps the first example of a triterpene benzoate isolated from nature [8,28].
Other compounds isolated from B. acutangula include tanginol [8,35, 36] as shown in FIG 7 and barrinic acid shown in FIG 8 (37].
Several compounds have been isolated, again from B. acutangute, and their structures assigned in part using NMR. These include barrigenlc acid, the 19G~isomer of barrinic acid (fruit) [36] and acutangulicand tangullc acids (leaves) (FIG 9) [38-40].
it was not until 1991 [41] that the structure of an intact saponin from B. acutangula was published. Spectral and chemical data led to the structure being assigned as 2a,3li,19a-trihydroxy-olean-12-ene-dloicadd28-CMJ.D-glucopyranoside (FIG 10) [41].
Shortly after the publication of this structure the same group published the complete structures of three more saponins from the seeds of B, acutangula, barrfngtosldes A, B and C (FIG 11) [42].
Although this Baningtonfa species has been used as medicinal plants for a wide variety of ailments, no information concerning the biological activity of any of the isolated triterpenes can be located. However, it is known that triterpenes have some anti-inflammatory activity (eg [43, 44]), The astringent properties of the bark of Barrlngtonia have been attributed to the presence of tannins [16,18] which are also known to possess anti-microbial

properties (eg [45]). The general use of Bam'ngtonia species as preparation for skin sores, wounds and other skin complaints may be due to the presence of these tannins. Many of the reported effects induced by preparations from these trees can be accounted for by the activity of steroids (eg anti-inflammatory, anti-asthmatic, anti-rheumatic etc) and the presence of the ft- and (- sitosterol and stigmasterol-3(J-OD-glucoside in extracts from Barringtonta species may explain some of these activities. It is also well known that saponins exhibit a wide range of biological activities, many of which could explain the observed medicinal properties outlined earlier (eg [46, 47]).
As Is evident from the preceding discussion, the dominant group of compounds found In Baningtonia thus far studies are saponins. Saponins are an Important class of secondary metabolites that are widespread in plants and lower marine organisms, ft has been reported that approximately 79% of all plants surveyed contain saponins [48], It has also been proposed that saponins are produced as defensive agents by the plant [48]. Increasing numbers of saponins are being isolated from lower marine organisms, but so far have been isolated from the phylum Ecbinodermata, in particular sea cucumbers (Holothuroidea) and starfish (Asteroidea) [46].
Saponins consist of three main components, an aglycone (genin or sapogenln), such as a triterpene, a steroid or a steroidal alkaloid, one or more sugar chains, commonly D-glucose, D-galactose, D-giucuronic acid, t>-galacturonic acid, L-rhamnose, L-arabinose, D-xylose and D-fucose and sometimes acids, such as angelic and tiglic acids [46,47,49], Saponins are further classified as mono-, bi- ortrkiesmosides according to the number of sugar chains which are attached to the aglycone [47].
The haemolytic, moUuscjdal and piscicidai activities of saponins are well characterised and have even been used as assay techniques In bio-guided fractionatbns of plant and animal extracts (eg [47, 48, 50, 51]). Howeverhaemolytic activity varies greatly or may be absent altogether and molluscicidal activity is somewhat dependent on the structure of the saponin [46, 47]. As expected there are many publications on the biological and

pharmacological properties of saponins, examples of which can be seen in [46,47J.
Analgesic activity has been demonstrated in a small number of saponins. The following is an example of some of the saponins found to have analgesic effects. Using the acetic acid writhing test It was shown that barbatoside A (ED50 95 mg/kg) and B (EDW 50mg/kg), glycoskJes of quaillic acid from Dianthus barbatus, were more active than acetyisalic acid (EDw 125mg/kg) {52]. An mtraperitoneal (/p) Injection of a saponin preparation from Dolichos falcatus at 5mg/kg was shown to produce marked analgesic effect to pain induced by exposure of 55°C in mice |53]. An extract of Platycodon grandiflorwr) also induced analgesia in mice when injected subcutaneous (sc) at a dose of 2g/kg [54]. One of the active ingredients is stated as being platicodin and the dose received by the mice was equivalent to 160 mg/kg. The effects were comparable to 100-200mg/kg aspirin. Injection (ip. 100-250mg/kg) of the total saponin preparation from Panax notoginseng was found to act faster but for shorter durations than morphine and Metrahydropalmatine and was comparable to amJnopyrine (160mg/kg) [56]. ft was also noted that the saponin preparation induced a sedative effect, decreased the ED50 of pentobarbital in sleep induction, prolonged thiopental induced sleep and showed synergfstjc effects with chlorpromazine in CNS Inhibition [55]. A number of dianosides were isolated and characterised from Dianthus superbus ver. Longlcafydnus [56^58]. Dianosides A and B were found to significantly inhibit acetic acid Induced writhing at 10 and 30mg/kg {sc) with dianoslde B the more potent [56], In a detailed examination of the pharmacological effects of glycosidal fraction obtained from Maesa chisa var. augustffolia, Gomes ef a/[59J demonstrated, among other things, analgesia in the writhing test in mice. A 33% inhibition was observed in p-phenylquinonone induced writhing In contrast to a 52% inhibition observed in acetic acid induced writhing. By comparison, aspirin inhibited p-phenylquinons Induced writhing by 85% and acetic acid induced writhing by 80%. The absence of straubtail phenomena and lack of activity in both the hot plate and the tail flick tests suggests that the analgesia produced by the

glycoside fraction was different to that produced by narcotics.
One aspect of the invention, and by no means the broadest form, provides for novel compounds of the formula (I)

R2 is selected from hydrogen, hydroxyl, O-alkyl, O-alkenyl, O-benzoyl, O-
alkanoyl, O-alkenoyl, O-aryl, O-heterocyclic, O-heteroaryl or
wherein R5 and R7 are "independently be selected from hydrogen, alkanoyl,
alkenoyl, benzoyl or benzoyl alkyl substituted alkanoyl;
R3 is selected from hydroxyi, O-alkanoyl, O-alkenoyl, O-benzoyl, O-alkyl, O-
alkenyl, O-aryl, O-heterocyclic or O-heteroaryl;
R4 is selected from -CH2OH, COOH, CH20COCH3( COO alkyl, COO aryl,
CH2COO alkyl. COO-heterocyclict COO-heteroaryl, CH2-O aryl, CH2O
heterocycfic or CH2O heteroaryl;
R6 is selected from hydrogen or

Amended Sheet IPEA/AU

Ri is selected from hydrogen or alkyl; or
pharmaceutically acceptable salts thereof, with the provisos that when;
R2 is OH, R3 is OH, R4 is CH2OH, and R8 is xyiopyranosyl, R1 is not H;
R4 is CH2OH and R3 is O-alkanoyl R2 is not O-acetyl;.
R4 is CH2OH and R2 is O-alkenoyl R3 is not hydroxyl; and
R4 is CH2OH and R3 is hydroxyl then R2 is not hydroxyl.
The term "alky!" refers to linear, branched, cyclic and bicyclic structures and combinations thereof, having 1 to 18 carbon atoms. Non-limiting examples of alkyl groups include methyl, ethyl, propyl, isopropyl, butyl, s- and t-butyl, pentyl, hexyl, heptyl, octyl, nonyl, undecyl, dodecyl, tridecyl, tetradecyl, pentadecyl, cyclopropyl, cyclobutyl, cyclopentyl. cydohexyl and the like. More preferably alkyl is selected from methyl, ethyl, propyl, isopropyl, butyl, s- and t- butyl, pentyl. and hexy!.
The term "alkenyl" refers to unsaturated linear or branched structures and combinations thereof, having 1 to 7 carbon atoms. Non-limiting examples of alkenyl groups include, ethenyl, propenyl, isopropenyl, butenyl, s~ and t-butenyl, pentenyl, hexenyl.
"Alkanoyl" means alkanoyl groups of a straight or branched configuration having 1-8 carbon atoms. Preferably alkanoyl is selected from acetyl, propionoyl, butyryl, isobutyryl, pentanoyl and hexanoyt. More preferable alkanoyl is selected from acetyl, propionoyl, butyryl, and isobutyryl.
"Alkenoyl" means alkenylcarbonyl in which alkenyl is as defined above. Preferably alkenoyl is selected from pentenoyl, hexenoyl or heptenoyl. More preferably alkenoyl is selected from petnenoyl (tigloyl) or hexenoyl (angeloyl).
The term "benzoyl alkyl substituted alkanoyl" is used to refer to straight or branched C1-C6 alkanoyl substituted with at least one benzoyl and at least one alkyl, wherein the benzoyl is attached to an straight or branched C1-6 alkyl. Preferably a benzoyl alkyl substituted alkanoyl is benzoyl methyl isobutanoyl.
"Heterocyclic" refers to a non-aromatic ring having 1 to 4 heteroatoms
Amended Sheet IPEA/AU

said ring being isolated or fused to a second ring selected from 3- to 7-membered alicyclic ring containing 0 to 4 heteroatoms, aryl and heteroaryl, wherein said heteroatoms are independently selected from O, N and S. Non-limiting examples of heterocyclic include pyrrolidinyl, piperidinyl, piperazinyl, morpholinyl, tetrahydrofuranyl, rmidazolinyl, thiomorpholinyl, and the like.
"Aryl" means a 6-14 membered carbocyclic aromatic ring system comprising 1-3 benzene rings. If two or more aromatic rings are present, then the rings are fused together, so that adjacent rings share a common bond. Examples include phenyl and naphthyi. The aryl group may be substituted with one or more substituents independently selected from halogen, alkyl or alkoxy.
The term "heteroaryl" as used herein represents a 5-10 membered aromatic ring system containing a single ring having 1-4 heteroatoms, selected from Of S and N. Heteroaryl includes, but is not limited to, furanyl, diazinyi, imidazolyl, isooxazolyl, isothiazolyl. pyridyl, pyrrolyl. thiazolyl triazinyl and the like.
Preferably R2 is hydrogen, O-benzoyl, O-tJgloyl, or
wherein R5 and R? are selected from hydrogen, tigloy(r benzoyl, or benzoy! alkyl substituted alkanoyl.
Preferably R3 is selected from hydroxyl, O-acetyl, O-benzoyl, O-isobutyryl or O-tigloyl.
Preferably R4 is selected from -CH2OHr O-acetyl or hydroxyl. Preferably the compound of formula (I) is selected from;
a. 3-0-p-D-xylopyranosyl(1^3)-[P-D-galactopyranosyl(1^2)]-p-D-
arabinopyranosyl]-22-0-acetyl barringtogenol C;
b. 3-0-p-D"Xylopyranosyl(1*>3HP-D-galactopyranosyl(1-^2)]-p-D-
glucuronopyranosyI-21-O-benzoyi barringtogenol C;
Amended Sheet IPEA/AU

c. 3-0-fi-D-xylopyranosy1(1 -»3H glucuronopyranosyl-21-C>-benzDyl-28-0-acetyf baningtogenol C;
d. 3-Ofi-D-xylopyrano8yJ(1 ->3H^D-galactopyranosyl(1 -^2)HH>
g(ucuronopyranosyl-21-O-ben2Dyl-22-CMsobutyryJ barringtogenol C;
e. 3-O-&-D-xylopyranosyl(1 ->3)-[&«D-ga!actopyranosyi(1 -*2)]-IJ-D-
methylgluc«tTonopyrano«y)-21,22-O-dlbenzoyl baningtogenol C;
f. 3^-firl>xyk5pyrano8yl(1^3Hfi-D-galactopyranosyK1-^2)]-erD-
glucuronopyranosyl-21.22-O-dibenzoyl barringtogenol C;
g. 3-O-fi-D-xylopyranosyl(1->3H^D-galactopyranosyl(1 •^2)]-R'D-
methylglucuronopyranosyl-21-0-benzoyl-22-0-tigloyl baningtogenoJ C;
h. 3-0-B-D-xy!opyranosyl(1 ->3H^l>9a'actopyrBnosyK1 ->2)]-&-l>
glucuronopyran08yl-21-0-benzoyJ-22-Ot^loyl bamngtogenoJ C;
I 3-0-fi-D-xylopyrano8y1{1 ->3HG-D^a'actopyranosyK1 -^2)J-R-D-
methy1giucuronopyranosyl-21,22-0-tigloyl barringtogenol C;
j. 3-O-R~D-xylopyranosy1(1 ->3)-[^D^alactopyranosyJ(1 ->2)J-D-
glucuronopyranosyl-21,22-0-tigloyl barringtogenol C;
k. 3-0-li-D-xylopyranosyl(1 -> 3MCrD^alactopyranosyt(1 -> 2)]-R-D-
glucuronopyranosy1-22-Oben2Dyl barringtogenol C;
I. 3~0-G-D-xylopyrariosyl(1->3HGD-ga!aotopyranosyK1 -^2)]-(i-D-
barringtogenol C;
m. 3-0-li-D-xylopyranosyl(1^3HG-D-gafactopyranosyK1 ^2)^6-D-
baningtogenol C;
n. 3-CM^D-xylopyranosyK1 ->3Hfr-D-galactopyranosyK1 -»2)]-R-D-
arabinopyranosyIl-22-O-acetyl baningtogenol C;
o. 3-0-G-D-xylopyranosyl(143HS-D-galactopyranosyl(1->2)]^r>
glucuronopyrarK3syl-21 arabinopyranosyl3-22-0-acetyl baningtogenol C;
p. 3-O-R-D-galactopymnosyK1 ->2HS-D-glucuronopyranosyf-21 -O-[3-(3-

barringtogenol C; or
q. 3-0-R-D-xyk>pyrano$yf(1 ^3H^aJaGtopynano5yl(1 ^2)}-QrQ-
arabinopyranosyq-28-O-acetyl barringtogenoi C.
The term *phamiaceutically acceptable salts: as used herein refers to salts which are toxicologlcally safe for systemic administration. The pharmaceutfcaHy acceptable salts may be selected from the group Including alkali and alkali earth, ammonium, aluminium, iron, amine, glucosam/ne, choline, sulphate, bisulphate, nitrate, citrate, tartrate, bitarate, phosphate, carbonate, bicarbonate, malate, maleate, napsylate, fumarate, succinatc. acetate, terephthafate, pamoate, pectinate and s-methyl methlonine salts piperazine and the like.
Another aspect of the invention resides in a composition containing one or more compounds of formula (I) when extracted from plants or parts of plants of the genus Barringtonla, preferably the species Barringtonia acutangufa. The parts of plants include fruit, seed, bark, leaf, flower, and wood. Preferably the part of plants is selected from bark, flower and leaf. More preferably the parts of plants is bark.
Another aspect of the invention resides in a pharmaceutical composition for treatment and/or control of pain comprising an effective amount of one or more compounds of formula (I) and a pharmaceuticaUy acceptable carrier.
Dosage forms include tablets, dispersions, suspensions, injections, solutions, syrups, troches, capsules, suppositories, aerosols, transdermal patches and the like. These dosage forms may also include injecting or implanting devices designed specifically for, or modified to, controlled release of the pharmaceutical composition. Controlled release of me therapeutic agent may be effected by coating the same, for example, with hydropnobic polymers including acrylic resins, waxes, higher aliphatic alcohols, poryactic and pofyglycofic acids and certain cellulose derivates such as hydroxypropylmemyl cellulose. In addition, the controlled release may be affected by using other polymer matrices, llposomes and/or microspheres.

Pharmaceutically acceptable carriers for systemic administration may also be incorporated Into the compositions of this invention.
Suitably, the pharmaceutical composition comprises a pharmaceutically-acccptable exoipient. By "pharrnaceutically-acceptable exdpient" is meant a solid or liquid filler, diluent or encapsulating substance that may be safely used In systemic administration. Depending upon the particular route of administration, a variety of earners, well known in the art may be used. These earners or exclpients may be selected from a group including sugars, starches, cellulose and its derivates, malt, gelatine, talc, calcium sulphate, vegetable oils, synthetic oils, polyols, alginic acid, phosphate buffered solutions, emuleifiers, isotonfc saline, and pyrogerbfree water.
Any suitable route of administration may be employed for providing a patient with the pharmaceutical composition of the invention. For example, oral, rectal, parenteral, sublingual, buccal, intravenous, intraarticular, intramuscular, intra-dermal, subcutaneous, inhalattonal, intraocular, intraperitoneal.intracerebroventricular, transdennal and the like may be employed.
Pharmaceutical compositions of the present invention suitable for administration may be presented in discrete units such as vials, capsules, sachets or tablets each containing a predetermined amount of one or more pharmaceutksally active compounds of the invention, as a powder or granules or as a solution or a suspension in an aqueous liquid, a non-aqueous liquid, an oil-in water emulsion or a water in oil emulsion. Such compositions may be prepared by any of the method of pharmacy but all methods include the step of bringing into association one or more pharmaceutically active compounds of the invention with the carrier which constitutes one ore more necessary ingredients. In general, the compositions are prepared by uniformly and intimately admixing the agents of the invention with liquid carriers or finely divided solid carriers or both, and then, if necessary, shaping the product in to the desired presentation.
The active compounds of formula (I) and of the composition of this

invention are present In an amount sufficient to treat and/or control pain. Suitable dosages of the compounds of formula (I) and the pharmaceutical compositions containing such may be readily determined by those skilled in the art but may be of the order of 0.002mg/kg to 5.0mg/kg.
In yet another aspect of the invention, there is provided a method of treating and/or controlling pain, comprising administering to a subject in need of such treatment an analgesteaHy effective amount of one or more compounds according to formula (I).
In yet another aspect of the invention, there is provided the use of one or more of the compounds according to formula (I) in the manufacture of a medicament for the treatment and/or control of pain.
FIG 1 shows the structure of A1 - barringenol:
FIG 2 shows the structure of barringtogenic acid and barringtogenol;
FIG 3 shows (a) initial and (b) revised structures of barringtogenol B;
FIG 4 shows structure of barringtogenol C;
FIG 5 shows structure of barringtogenoJ D;
FIG 6 shows structure of barringtogenol E;
FIG 7 shows compounds from B. acutangute:
FIG 8 shows structure of barrinic acid;
PIG 9 shows compounds from B acutanoula including acutangulic acid, tangulic acid and barringenlc acid;
FIG 10 shows structure of 2a, 3fi, 19a tri hydroxy-olean-12-ene dloc acid 28 - 0 - ft - D glucopyranoside;
FIG 11 shows structure of barringtoslde A, barringtosfde B and baningtoside C;
FIG 12 shows normal grooming response as control in Formalin assay;
FIG 13 shows control values in formalin assay (x ± S.E; n = 2); FIG 14 shows dose control curve for morphine (x ± S. E; n = 6 (min)]; FIG 15 shows schematic for preparation of crude saponin mixtures;

FIG 16 shows acid and base hydrolysis scheme for insoluble active portion of water extract;
FIG 17 shows analgesic activity of flowers and leaves ofBaoetanaula (x ± S.E; n - 2);
FIG 18 shows analgesic activity of crude water extract (x ± S.E; n -
FIG 19 shows analgesic activity of crude water soluble (n = 9) and insoluble (n = 4) portions of the water extract (x ± S.E);
FIG 20 shows dose response curves for water extract (x ± S.E; n =
FIG 21 shows preparative gel permeation column;
FIG 22 shows dose response curve for TSK - 4a (x £ S.E; n - 3); FIG 23 shows C18 separation of TSK - 4a; FIG 24 shows C18 preparative separation of TSK - 4a; FIG 25 shows preparative C18 chromatogram of HzO extract; FIG 26 shows outline of numbering system in regard to various fractions for compound F70.2.53;
FIG 27 shows separation of fraction elating at 70% MeOH (F70); FIG 28 shows separation of fraction 70.2 (40% MeCN In 1% TFA); FIG 29 shows chromatogram of F.70.2.6; FIG 30 shows separation of fraction F. 70-2.2 at 264nm (left) and 233nm (right);
FIG 31 shows separation of fraction F70.2.5 at 220nm (left) and 233nm (right);
FIG 32 shows separation of fraction F70.3; FIG 33 shows chromatograms of F.70.3.5 and F70.3.7; FIG 34 shows analytical separation of fraction F. 70.3.4 (predominantly single compound);
FIG 35 shows separation of F70.4;
FIG 36 shows separation of F70.4.2;
FIG 37 shows separation of F70.4.3;
FIG 38 shows preparative chromatograms showing loss of peaks

F.80.2 and F.80.3;
FIG 39 dhows preparative chromatograms of F.8G.4;
FIG 40 shows separation of fraction F.80.6 using a phenyi reverse
FIG 41 shows TLC plates used in hydroJyste procedure showing standard sugars for isolation and structural elucidation of F.70,3.6; FIG 42 shows UV spectrum of F.70.3,6; RG 43 shows FRIR spectrum of F.70.3.6; FIG 44 shows 1H - NMR for compound F.70.3.6; FIG 45 shows 1SC - NMR for compound FJQ.3.6; FIG 46 shows the complete assignment of structure to compound F.70.3.6;
FIG 47 shows negative ion HR-ESMS of F.70.3.6; FIG 48 shows compound F.70-2.52; FIG 49 shows compound F.70.2.3; FIG 50 shows compound F.70.3-2; FIG 51 shows compound F.7G.3.4.2; FIG 52 shows compound F.; FIG 63 shows compound F.80.8.4/F.; FIG 54 shows compound F.; FIG 55 shows compound F.; FIG 56 shows compound F.70A.322; FIG 57 shows compound F,80,6,2; FIG 58 shows compound F.; FIG 59 shows compound F.; FIG 60 shows compound F.; FIG 61 shows compound F.70.3.5a; FIG 62 shows compound F. 70.3.5b; RG 63 shows compound F.; FIG 64 shows compound F. F.80.5.2; FIG 65 is a graph of the mean (± SEM) paw withdrawal threshold versus tfm© curves for (A) ipsilateral (inftemed) and (B) contralateral {non-

inflamed) hindpavys of FCA-rats;
FIG 66 Is a graph of the mean (± SEM) paw withdrawal threshold versus time curves for the (A) ipslteteral (Inflamed) and the (B) contralateral (non-inflamed) hindpaws of FCA-rats;
FIG 67 is the mean (± SEM) paw withdrawal threshold versus time curve for the ipsilateral (inflamed) and the contralateral (non-inflamed) hindpaw in FCA-treated adult male Sprague-Dawtey rats (n = 3) that received a single i.v. bolus of saline;
FIG 68 is the mean (± SEM) dose-response curves for the antinociceptive effects of I.v. bolus doses of F70.3.2 and F70.3.6 in the ipsHateral hindpaws of FCA-rats: and
FIG 69 is a graph of paw volume pre and post FCA treatment
The formalin assay involves the subcutaneous injection of a small amount of formalin into the fore or hind paw of a rat or mouse and the behavioural response to this injection is recorded as a measure of pain response. A modification of this method was described by Dubuisson and Dennis [60] and the behavioural response detailed for both rats and cats. Independently the pain produced was described as being initially intense, sharp, stinging and burning and was given a 3/5 on a standard pain questionnaire. Some 4 to 5 minutes later this intense pain gave way to a steady throbbing ache which gradually disappeared over 30 to 60 minutes leaving a mild tenderness at the injection site [60].
The formalin assay was chosen for the current work for several reasons. Firstly, and most importantly, this assay is often reported in Journals such as Pain, which would indicate that ethical considerations have been overcome. It has also been demonstrated that the two distinct phases observed in the assay reflect two distinct phases of nociception. The first phase, early or acute, begins immediately after injection of formalin and lasts

some 3-5 minutes. This phase is considered to be the direct chemical stimulation of nociceptors. A period of minimal activity lasting for 10 -15 minutes follows this initial phase. Subsequently a second, late or tonic phase, begins and lasts for 20 - 40 minutes. The response shown in both early and late phases can be reduced using known analgesics, such as morphine, codeine, nefopam and orphendarine [81]. The late phase was affected both by non-steroidal anti-inflammatory compounds, such as indomethacin and naproxen and steroids such as dexamethasone and hydrocortisone (e.g. [61]). Interestingly aspirin and paracetamol were shown to be analgesic in a dose dependent manner in both phases of the formalin test [61].
When performing this assay several points, including formalin concentration, experimental subject and site of injection need to be considered.
The formalin assay was preformed on male mice using morphine, and other compounds of Interest, as an analgesic, to test the viability and reliability of the assay as a means of measuring pain response In mice. Methods and materials. Experimental subjects.
Mate Quackenbush mice weighing 25 - 35g ware used. They were housed in colony cages (400 x 300 x 130mm; Wlretainers) with ad llbttum access to food (standard rat/mouse pellets; Norco Feeds) and water. For short term storage a maximum of 15 -16 mice were housed per cage while for longer periods this number was reduced to less than 10 mice per cage. Bedding material was wood shavings or more commonly recycled paper pellets (Breeders Choice).
A12 hour light/dark cycle was maintained in the animal holding fecHity with lights on at 06:00hrs, All testing was performed during the light phase to minimise any diurnal variation in behaviour. The temperature of the fadlfty was maintained at 21 °C with humidity ranging between 45 and 65%.
Newly acquired animate were housed in the holding facility for a minimum of two days prior to being used for testing.

Test compounds.
Formalin (Ajax) was supplied as solution of approximately 37.5% formaldehyde containing 10% methanol as a preservative. This stock solution was diluted 12.0 with water to grve a 2% formaldehyde solution (5% formalin) which was used as the nociceptive agent.
Morphine hydrochloride was kindly donated by Extal, a division of Tasmanian Alkaloids Pty. Ltd. and appropriate concentrations were made by dilution with sterile isotonic (0.9%) saline. Testing method.
In the absence of a dedicated room for testing, aH tests were performed in the laboratory on weekends during times consistent with the first hours of the light phase In the holding facility. These times were chosen to minimise disturbances during testing due to other workers in the laboratory and to minimise any variation in response due to time of the day.
The mice were brought into the laboratory at least one hour prior to testing. Subsequently the mice were individually placed into empty colony cages, which also served as observation chambers. They were allowed a further 30 minutes to explore their environment. During this time, and during testing, no food or water was available.
Morphine and other compounds of interest were injected in a volume of 10mL/kg Intraperftoneally(ip) 30 minutes prior to formalin administration. With minimal restraint 20uL of formalin solution was Injected subcutaneousiy (sc) into the dorsal surface of the right hind paw. The mouse was replaced into the cage and observations begun immediately. The sole behavioural response recorded was the amount of time the animal spent biting or licking the injected paw or leg per 5 minute block was recorded. Each animal was used once only for testing and were subsequently euthanased by CO2. Results. - Control.
Two control experiments were performed to ensure that the assay gave comparable results to the literature. In addition a background, grooming behaviour was recorded. This data was recorded as the amount of time that the mouse spent biting or BcWng both hind paws, with the average value

taken as the normal grooming background (FIG 12). The mouse was given \p saline only for these observations.
FIG 12 indicates that the amount of time that the mice spend grooming their hind paws is approximately 4 seconds every 5 minutes. Although there was some small variation in this time, ft was decided that these values were hot significant and were therefore not Included in further calculations.
The second control experiment involved an ip injection of isotonic saline followed, 30 minutes later, by injection of formalin. The results for the controls are shown in FIG 13. The characteristic Diphasic nature of the response reported by previous workers (e.g. [61]) is shown in FIG 13.
The final control experiment conducted to ensure the validity of the technique was to construct a dose response curve for morphine. Four doses of morphine were chosen, 3mg/kg (n=8), 6mg/kg (n=7), 9mg/kg (n=6) and 12mg/kg (n=4). A calculation of analgesic activity over the total experimental time period was required to evaluate the extracts as analgesics. Any differences observed between the acute and tonic phases of the nociceptive response were noted. A comparison between the duration of experimental and control pain for the entire experimental period (45min) was needed.
The total time spent exhibiting the pain response In control animals was 451 ± 28 seconds (x ± S.E.; n = 18). Using this value a percentage pain inhibition can be calculated using the formiia:-
Pain response (%) * (1 - Response time(sec)/Cortfrol time) *100
-(1~ Response t}me($ec)/451) * 100
where response time is the average pain response time recorded over 45 minutes when observed after administration of the compound of Interest. The morphine dose response curve for the entire experimental period is shown in FIG 14 (ED50 - 4.8 mg/kg).
Previously it has been demonstrated that-the effect of morphine is greater in the later, tonic phase than in the eariy, acute phase in the fbrmatln

assay. An approximate EDgo value of morphine between 4 to 5 mfl/kg has
been reported for the acute phase in mice whereas in the tonic phase 2 to 3 mg/kg of morphine ia required (e.g. (62]). For the entire experimental period subcutaneous morphine gave an ED50 value of 4.8 mg/kg [63] and It has
also been noted that almost complete analgesia was induced at 6 mg/kg [64]. The results obtained for the current work showed similar trends. Summnry.
In summary, the formalin assay was found to be reliable, easy to perform and provided sufficient information from a minimal number of experimental subjects. The control results of the assay compare favourably with those of other workers. The method conforms to the ethical guidelines of the ISAP (65, 66] and the Griffith University Ethics Committee for Experimentation on Animals (GUECEA).
Crude extracts and tractions from Barringtonla were tested for
effectiveness in pain inhibition using the mouse formalin assay.
The separation of saponins, and subsequent characterisation, requires the use of sophisticated techniques. Purification of saponins is achieved by many methods that are discussed later (e.g. [67-60]), but in general involves the following (FIG 15). Extraction in aqueous alcohol (methanol or ethanof), either preceded or followed by a defatting step, removal of the solvent and suspension of the residue into water saturated n-butancl. At this stage the saponins can be precipitated with dtethyl ether, however this step may be omitted. The residue Is subsequently subjected to chromatography. This last step may involve several techniques, including Sephadex LH20, silica, Diol or reverse phase (C8 and C18) cofomatography, alone or in combination. The technique of counter current chromatography and Its variations (DCCC, RLCC, CPC) has ateo found application in the separation of saponins.

The major problem associated with the isolation of saponins using chromatographlc techniques is the lack of a suitable chromophore for UV detection. Although these problems can be overcome by using Rl» mass detection and by derivatisation and UV detection, each of these techniques has its own inherent advantages and difficulties which have been discussed elsewhere [67],
The use of bioassay to guide the purification may, however, dictate the Isolation methods used, This section describes the processes which led to the separation of several pure saponins from the aqueous extract of the bark of Barringtonia acutangula. Experimental. General methods after materials.
The mouse formalin assay was performed as described previously. Assay results were reported as the percentage inhibition of the pain response compared to controls (see Section A).
Unless stated otherwise all solvents and reagents were AR grade. Pure water (ddHzO) (Permutit Australia, conductivity were filtered (0.4SJ m nylon, Actlvon) before use.
NMR was performed using a Varian 200MHz (Gemini 200), 400MHz (Unity), 500mHz (Unity Inova) or600MHz (Unity Inova fitted with Ultra Shims) system using deuterated solvents (CD3OD, CD6N, (fe-DMSO). Where only small amounts of sample were available, D2O or DMSO matched Shigemi NMR tubes (3 or 5 mm) were used in order to present a more concentrated sample to the NMR. All chemical shifts were reported in parts per million (ppm). Standard Varian pulse sequences were used for all experiments.
Low resolution electrospray mass spectrometry (LR-£SMS) was performed on a Rsons VG Platform LCMS connected to a Waters 600 HPLC

system (methanol, flow rate 0,9mL/min). Spectra were collected in both negative and positive ion modes, at a range of cone voltages (±50V, ±100V, ±150V and ±200V), and were analysed using MassLynx software. High resolution ESMS (HR-ESMS) was performed at the Australian Institute of Marine Science, Townsville Queensland. The instalment was a Bruker (Billerica MA, USA) BioApex 47e FTMS equipped with an Analytica of Branford (Branford CT, USA) external etectrospray source, .The Instrument was calibrated in either positive or negative mode prior to sample Infusion and molecular masses were reported to within 5ppm. Plant material.
The plant sample was identified as Bam'ngtonm acutangula (L) Gaetm. ssp. acutanguia and a voucher was lodged at the Queensland Herbarium (#AQ595351). Although the tree is found across Northern Australia, the sample used in this project was collected from the Wmbertey district of North Western Australia. An initial collection of the bark from the tree was made on July 14th 1989 and a subsequent collection was made in 1994. In order to ensure simitar characteristics the collection was rnado from the same area and at the same time of tine year (July 18* 1994). A small sample of flowers and leaves was also made available.
The flowers of B, acutangula are small and a large number were required to provide sufficient material for further investigation. The tree flowers In the rainy season, at which time the area becomes flooded making collection difficult Therefore no further collections were possible and the activity in the flowers remains to be investigated. A quantity of bark was available from the initiaf collection from which the activity had been demonstrated. Therefore the current work aimed to characterize the analgesic activity in the bark.
Bark was removed from the trees by the aboriginal peopte who lived in the area in such a manner as to inflict minimal damage and ensure the continued growth of the tree. The bark samples were air or oven dried at a
maximum temperature of 50°C, The dried samples were mill ground to a coarse powder and the powdered sample stored in an airtight container at

room temperature until ready for extraction.
The small sample, flowers and (eaves were dried, powdered and stored in a similar manner the bark. Extraction of plant material.
The dried and powdered bark was soaked for several hours In approximately ten volumes (w/v) of deminerallsed water (dHaO). The resulting mixture was filtered through several layers of muslin, centrifuged {Damon (EC Division DPR6000, 4500g for 45min) and the supernatant freeze dried (virtls Freezemobile 12). The dried extract was stored in airtight containers at 4°C.
The flowers (0.25g) and leaves (1.35g) were extracted with dhfcO In a similar fashion to the aqueous extraction of the bark.
Several methods were employed to fractionate the bark H2O extract and are described below. Method 1.
The water extract was redissolved in ddH^O for assay or further purification. A significant amount (~30%) remained undissolved and was removed by centrifugatron (Sorvall RC5B Plus, 12,000g, 20m(n) followed by filtration (0*45:m nylon, Activon). Where required the H2O insoluble portion of the extract was assayed as a suspension.
In addition to H2O, several alternative solvents were used and the filtrate (0.45:m nylon, Activon) from these extractions was subsequently assayed. Alternative solvents included MeOH, CHCfe, MeOH:CHC!a (1:9), MeOHirifluoroacetic acid (TFA, 0.5%), ethyl acetate (EtOAc). 1% ammonia solution (NH4OH), dimethyl sujfoxide (DMSO) and dimethyl fomnamide (DMF). Acetylation, hydrolysis and sonication of the H2O insoluble portion are described below. Extracts were assayed as described previously. Method Z
Dried water extract, prepared as in method 1, was dissolved in 1% ammonia solution (AJax) and centrifuged (Sorvall RC5B Plus, 12,000gr 20min) to remove any insoluble material (~22%). The supernatant was applied to a TSKHW40S column (vide infra) and eluted at 5mL/mfn with 1 %

NHtOH. Initially water was used as the mobile phase, however the small amount of ammonia was found to produce sharper, better resolved peaks. The water extract was dissolved in the mobile phase to minimise any problems resulting from precjpitation of extract on the column. Initially five fractions, labeled as TSK-1, TSK-2, TSK-3, TSK-4 and TSK-5 were collected according to the resulting chromatogram. However, difficulties in consistently separating TSK-4 from TSK-5 led to both of these fractions being pooled as TSK-4a. Extracts were assayed as described previously.
The active fraction from the gel permeation (TSK-4a) column was dissolved in MeOH, filtered (0.45:m nylon, Activon) and applied to a preparative C18 column (vtofe Infra). The column was eluted using a step gradient of MeOH in ddHaO (0,35, 70 and 100% MeOH) and the fractions collected according to the resultant chromatogram. Method 3.
The method utilised in this section was similar in all respects to method 2, the only difference was in the step gradient used for preparative C18 chromatography. In this instance a step gradient was employed which consisted of 10% increments of MeOH in ddHzO (0-100% MeOH). Method 4.
In this method the water extract was applied directly to a C18 preparative column omitting the need for a gel permeation step. The water extract was dissolved in dH^A centrifuged (Sorval RC5B Plus, 12000g for 20mln) and the supernatant applied a C18 prepaistive column (vide infra) and the column eluted as in method 3. Separation of active fractions.
The methods used for separation of crude, active fractions Into pure compounds were developed on an individual basis. The columns used to facilitate separations included C18, C8, diol and phenyl bonded silica. Suitable mobile phases Included MeOH, MeCN, isopropanol (/-PrOH), hexane, EtOAc, HA 0.01 M HAc and 1 % TFA. In general the approach used was to begin with a C8 or C18 column and a gradient of MeOH or MeCN with H2O or 1 % TFA (early separations used 0.01 M HAc). Modifications to the

method were made until it became necessary to use a different column (e.g.
pheny! or dio!) and the process repeated in order to provide maximal
separation and resolution. The methods adopted for each fraction are
discussed in the relevant areas of the results and discussion sections.
Analytical and semi-preparative chromatography.
Semf-preparative and analytical chromatography were performed using a Waters 600 HPLC system fitted with a photodiod© array detector (PDA 996), sn autosampler (717 plus) and a fraction collector. Chromatographic information was collected and stored using Millennium 2010 chromatography manager software {version 2.10).
Analytical chromatography was performed using Dynamax columns in one of two formats. Reverse phase (C18 or phenyi) columns were efiher 4.6 X 250mm (8:m irregular silica, 60 A pore stee, imL/min) or 4,6 x 50mm ("Short Ones", 3:m spherical silica, 60 A pore size, 1 mL/min). Similarly, semi-preparative chromatography was performed using reverse phase (C18 or phenyl) columns which were either 10 x 250mm (8:m irregular silica, 60 A pore size, 4mL/min) or 10 x 50mm ("Short Ones', 3:m irregular silica, 60 A pore size, 4mUmln), All solvents used for analytical and semi-preparative chromatography were degassed by helium sparging. Preparative chromatography.
Large scale preparative chromatography (gel permeation and CIS-silica) was performed using either a Giison (Model 303 pump, 804 manometric module) or a Waters 600 HPLC system fitted with an extended flow kit. Both systems were connected to a UV detector (254nm, absorbanee mode) (ERMA Optical Works, ERC 7215), a fraction collector (ISCO Foxy or Waters) and either a chart recorder (Omniscribe series D5000) or an integrator (C-R3A Chromatopac, Shimadzu).
Gel permeation chromatography was performed using TSKHW40S packing (Merck) in a 49 x 350mm glass column (Buchi) (flow rate 5mL/min, 4m(n fractions),
C18-siJica preparative chromatography was performed using a 28 x

230mm glass column (BOchi) (flow rate 12.5mL/miny and fractions were
collected according to the resultant chromatogram. During the early stages of
the work separations were accomplished using laboratory prepared C18-
silica packing, in contrast to later separations where C18-Davis}l (ABtech)
packing material was used. The preparation of the C18 silica is described
Preparation of C18-sillca for early preparative chromatography.
The entire procedure was conducted under a dry nitrogen atmosphere. Toluene (AR, Ajax) was dried over sodium wire, distilled and stored over sodium wire and 3A sieves (Ajax). MeOH and dichloromethane (DCM) were distilled and stored over 3A sieves. A sfurry was made using dry toluene (400mL) and dry silica gel (200g, 40-63:m, Merck) to which octadecyl trfcfrlorosilane (40mL, Fluka) was added and the mixture stirred for 24hr. The slurry was subsequently filtered via a BOchner funnel and the product washed sequentially with 400ml each of dry toluene, dry MeOH and dry DCM. Finally the non-end capped C18-silica product was dried at 37°C for 24hr. End capping was achieved by suspending the non-end capped CIS-silica in 400mL of dry toluene, adding 40mL of trimethytahlorosHane (Fluka) and allowing the mixture to stand for 24hr. The final product was washed successively with 400mL dry toluene and 400mL dry DCM before being dried at 37°C for a further 24hr. Acetyiatlon of the Insoluble component of the water extract
A 1g sample of the water insoluble portion of the water extract was added to 5mL dry pyridine (reflux over sodium wire, store over 4A sieves). The mixture was cooled to 0°C> 1mU of acetic anhydride added and the mixture stirred overnight at room temperature (RT). Ice (1-2g) was added to quench the reaction. The mixture was partitioned into ethyl acetate (25mL of ddH2O and 25mL ethyl acetate). The organic layer was retained and washed successively with 10mL each of 1M HCI, ddhfeQ* saturated NaHCOg and again with ddH2O. Finally the organic layer was dried over NasSCU or MgSC>4 and the solvent evaporated to dryness.

Base hydrolysis of the Insoluble component of the water extract
A schematic forthte process is shown in Figure 16. A sample (0.5g) of the active, water insoluble, portion of the water extract was added to 20mL of 0.5M NaOH, stirred for 1 hr and subsequently centrtfuged (12,000g, 15min). The pellet (360mg) was washed repeatedly with ddHaO until the fiHraie became clear. A portion of this pellet (330mg) was extracted with 20mL of DCM and centrifuged as before to recover the pellet (320mg). Th© peflet (100mg) was added to 10mL of 2M NaOH, refluxed for 2hrs and allowed to cool overnight The mixture was partitioned successively into DCM (20mL) and n-butanol (20mL) and the solvent evaporated in vacuo. The pH of the aqueous layer adjusted to 7 with 1M HCI. Again the aqueous phase was successively partitioned into DCM and o-butanol and the solvent removed in vacuo. Finally the aqueous phase was acidified (pH 1) with 1M HCI, partitioned into DCM and n-butanoJ end the solvent removed as previously. The solvent remaining in the aqueous phase was also removed in vacua. Samples from all steps were assayed for activity. Arid hydrolysis of the Insoluble component of 0i« water extract
The procedure for addle hydrolysis ctosofy follows the procedure outlined for base hydrolysis. The active, water insoluble, portion of the water extractwas treated as forbase hydrolysis and a portion of the peBet (100mg)
was subjected to eofd hydrolysis ae outlined in FIG 16.
Sonicatton of the Insoluble component of the water extract
A sample of the active insoluble material was sonicated (20kHz,
300W for 10min) in water. Both the soluble and Insoluble material were
Results and Discussion,.
The aim was to examine analgesic effects of a water extract and
subsequently purify and characterized the compound(s) responsible for the
Flowers and leaves.
Aqueous extratlon of the flowers and leaves resulted in 12.8mg and
139.6mg of flower and leaf extract, respectively, which were used in a

formalin assay, as described previously. The results of the assays are shown
in FIG 17.
The activity of the flowers was very high (-70% Inhibition) at 5 mg/kg whereas the leaves produced little activity (-18% Inhibition) at the same dose. Method 1.
The powdered baric was extracted In dHaO to give a crude water extract (9.8%). The crude water extract was dissolved in CWH2O at a concentration of 10 mg/mL/kg and assayed for analgesic activity as previously described. These results can be seen in RG18.
From FIG 18 It was clear that this extract produced a considerable reduction in the pain response (78%) when compared to controls (see Section A). It can also be seen that the pain response, as measured by the formalin assay, is much reduced and confined to the acute stage. However It was noted that this extract contained paniculate matter, which was approx 40% of the total weight of the extract. This suspension was filtered and both the soluble and insoluble material assayed separately at a dose of 10 mg/mUkg. It was found that the analgesic activity was higher (66%) in the insoluble material than in the soluble material (45%). However in both cases some pain response was observed in the tonic phase of the assay, although this was more evident in the water soluble portion of the extract (FIG 18).
Acetylation of the active water Insoluble portion of the water extract reduced the activity from 66% to 28% pain inhibition and therefore dfd not prove to be a useful technique in the separation of active constituents.
Given these results it appeared that a significant amount of the activity was not soluble in water. An attempt was made to extract the activity from the insoluble component bark by using a variety of solvents (Table 1). AH extracts were assayed at 10 mg/kg/mL as previously described.
The Merck Index states that DMSO has antMnftemmatory activity, has been proposed as an analgesic and has also been used as a penetrant carrier to enhance absorption of compounds. Although the DMSO extracts

were freeze dried some residual DMSO always remained. Therefore it was important to determine whether residual DMSO interfered with the assay. Although a high dose was used (5%), DMSO was found to reduce the pain response by 57%, Therefore in order to avoid any interference, DMSO was not considered for any further extraction. Therefore, as can be seen from Table 1, the water extract was at least as active in the mouse formalin assay as the extracts from other solvents. As no additional extraction could be achieved by the other solvents It was decided to characterise analgesic activity In the water extract.
At this time it was decided to construct a dose response curve for both the water soluble and the water Insoluble extract and compare the results (FIG 20). From these results it can be seen that the amount of extract required to reduce the pain response by 50% (ED^) is approximately 36
mg/kg (soluble) and 5 mg/kg (insoluble). These results, combined with the ease of extraction, further supported the decision to concentrate further purification effort on the water soluble material.
Method 2,
This method involved a preparative gel permeation separation of the water soluble extract. Initially five fractions were collected and, with respect to the starting HzQ extract, the following average yields were obtained, TSK-1 0.4%, TSK-215%, T3K-3 23%, TSK-4 0.7%, TSK-5 20% and an Insoluble portion (22%). Fractions TSK1 to 5 were assayed at a crude water soluble equivalent dose of 100 mg/kg {i.e. approximately 2.5 times the EDgo) and the results can be seen in Table 2.
Although the greatest apparent reduction in the pain response was seen in fraction 2 (74%), this required 15.8 mg/kg of material, In contrast 0.7 mg/kg of fraction 4 was required to produce a 31 % inhibition of the pain response, suggesting that fraction 4 was approximately ten times as potent as fraction 2. However, the small peak which eluted as TSK-4 was often obscured by the much more intense peak of TSK-5. Therefore the separation of fractions TSK-4 from TSK-5 proved to be inconsistent and in subsequent

separations they were pooled and named TSK-4a. The following average yields, with respect to the starting H2O extract, were obtained, TSK-11.7%, TSK-2 23%, TSK-3 20%, TSK-4a 16% and an insoluble portion (17%) (FIG 21).
At this point a dose response curve was constructed for fraction TSK-4a (FIG 22). This graph showed that an ED50 for this extract was approximately 1.8mg/kg. This represents an overall purification of the analgesic activity some 30 times that of a crude water-soluble extract of the bark.
Further separation of the active fraction (TSK-4a) was achieved by
elutfon from a preparative C18 column with a step gradient of 0, 35,70 and 100% MeOH in H2O. The resulting chromatogram is shown in EJG 23.
The fractions were collected as evidenced by the chromatogram, aned in vacua and assayed. The following average yields, with respect to the TSK-4a starting material, were obtained, 0%-MeOH 64%. 35%-MeOH 17%, 70%-MeOH 2.4%, 100%-MeOH 0.7% and an Insoluble portion 2.2% The dosage used for assay was an equivalent to 3 mg/kg of the TSK-4a (approx two times ED50)- The results of the assays are show in Table 3.
ThPrtw rttaMH* «h«wed aetivftv in all fractions eluted from the column, although the activity eluted at 0% was small in comparison to the other
fractions. (It was of interest to note that the fraction which eluted In 70% methanol sedated the mice whereas the other fractions did not). Attempts to further separate each fraction indicated that, as expected, each fraction still contained a large number of compounds. Method 3.
Using a preparative C18 silica column to separate the activity demonstrated in fraction TSK-4a, the number of steps, in the gradient was increased from four to eleven (10% Increments, 0-100% MeOH in H2O). The resulting chromatogram can be seen in FIG 24 and the yields obtained, with respect to the TSK-4a starting material, were 0%-MeOH (F0) 61.5%, 10%-MeOH (F10) 3.6%, 20%-MeOH (F20) 3.2%, 30%-MeOH (F30) 2.9%, 40%-

'MeOH (F40) 2.6%, 50%-MeOH (F60) 1.6%, 60%-MeOH (F60) 0.9%. 70%-MeOH (F70) 2.4%, 80%-MeOH (FBO) 0.2%. 90%-MeOH (F90) 0.1%. 100%-MeOH (F100) 0.2% and an Insoluble portion 3.1%.
Again each fraction was assayed and the results can be seen in Table 4. The results of these assays indicated that the most potent fractions, on a % inhibftion/mg of extract basis, were those eluted at 70.80 and 90% MeOH.
However, it soon became obvious that each of these fractions still contained a large number of compounds and that the yields obtained via this method were insufficient to supply sufficient pure compound for further analysis. It was suspected that cumulative losses during the isolation steps resulted in the low yields. Therefore large scale work-up by direct chromatography on C13 was performed in order to obtain large quantities for further fractionation.
It was established that most activity was eluted from a C18 column using 70, 80 and 90% MeOH as the mobile phase. Therefore the gel permeation step was omitted and the water soluble extract was applied directly to a C18 preparative column after removal of any insoluble portion by centrifugation and filtration. The resulting chromatogram is shown In FIG 25.
The first of the separations which were performed using this method used C18 silica which was prepared in the laboratory whereas later separations were performed using commercially available C18 packing material (ARtech Davtell). A comparison between the yields obtained using both packing materials is shown fn Table 5 below.
As had been previously decided, In order to minimize the number of animals used no further testing for analgesic activity was performed. Therefore activity in each of ttie fractions which were eluted from this procedure was not confirmed. However, It should be noted that if activity was present in fractions which eJute from a C18 column at 70,80 and 90% MeOH in HA then it follows that the active constituents wffl also be found m the same fractions, albeit in lower concentrations, even though prior separation

steps were omitted, It was on this basis that the investigation continued to examine those fractions which eluted with 70, 80 and 90% MeOH in H2O from the preparative C18 column. A preliminary 1H-NMR of all fractions suggested that the predominant compounds in the fractions collected were saponins and tannins. The tannins were found in the fractions elutlng at lower MeOH concentrations while the saponins tended to elute at higher
«*"«" xHt-w-wtitraHnns from the C18 column. Initially it was decided to limit this Investigation to those fractions which eluted with 70,80 and 90% MeOH.
However it became apparent that a large number of compounds existed In
each of these fractions.
Numbering scheme for collected fractions.
In order to consistently identify fractions and compounds collected In the current investigation, the following numbering scheme was adopted. The first three digits identify the percentage methanol that eluted the original fraction from a C18 preparative column. The remaining digits identify the collected peaks in subsequent fractionations. This is shown diagrarnmaticaJIy in FIG 26 for fraction F70.2.5.2.

Fraction elutlng at 70% methanof (F70).
The fraction which eluted at 70% MeOH (F70) was further separated on a C18 semi-preparative (25cm) column using a MeOH:1 % trifluoroacetic acid (TFA) gradient The TFA was added to the mobile phase in order to sharpen the peaks In the chrornatogram. Earty separations in the current project, which used single wavelength detectors and chart recorders, used acetic acid (0.01 M) to improve the resolution. However TFA was chosen for later separations after it was determined that residual acetic acid was more difficult to remove than residual TFA. It was at this point that the advantages of using a photodiode array detector (PDA) became apparent Conventional UV detectors are usually able to monitor the separation at a single wavelength. However the PDA was able to monitor the separation at several wavelengths simultaneously. As an example consider the chromatograms shown in FIG 27. In general, saponins have no UV chromophore and are more likely to be detected at wavelengths approaching 210nm. However it

was found that high absorbances could be achieved at 233nm. Indeed a separation of the extract couW be adequately observed at 254nm indicating the saponins present in the extract contained UV active chromophores. FIG 26 shows that, at the conditions employed tor the separation and employing the three wavelengths shown, the extract could be separated into 5 distinct fractions. Varying the gradient conditions did not result In further separation of the peaks and therefore these five peaks were collected as indicated for further purification steps. Table 6 presents the yields obtained and preliminary 1H-NMR results which Indicate that no pure compounds were isolated at this point fraction F70.2.
Fraction F70.1 was not further investigated as 1H-NMR suggested that a considerable carryover from fraction F60 was present. Attempts to separate F70.2 (C18 25cm) by altering the percentage of MeOH in the mobile phase did not result in further separation of fraction F70.2. A change in mobile phase to acetonitrile (MeCN):1% TFA mixtures proved more successful and an isocratic run of 40% MeCN in 1 %TFA resulted in the separation of this fraction into 8 peaks (FIG 28 and Table 7).
Preliminary 1H-NMR showed that fractions F70.2.3, F70.2.4, F70.2.6 and F70.2.7 were predominantly single compounds and repeated chromatography using identical conditions resulted in pure compounds with sufficient material to begin structural elucidation. However, on standing, the NMR samples In ag-DMSO were found to contain minor Impurities. This suggested that some minor hydrolysis of the compounds was occurring, possibly due to residual TFA which was not completely removed on drying. As the small amount of impurity present did not interfere with the structural elucidation, and the amount of pure compound was small, obtained no further attempts to remove the impurity were made. A chromatogram showing chromatographic purity of fraction F70.2.6 recorded at 220,233 and 254nm is shown in FIG 29.
As previously mentioned purity of ail fractions was determined by 1H-NMR, the NMR solvent being aVpyridine. However, when ihe solvent was

removed by freeze drying, one or more compounds within the fraction were found to have reacted with the NMR solvent to produce a bright pink complex. This was particularly noticeable with fraction F70.2.2.1H-NMR of the complex tended to suggest that the compounds in the fraction had not altered, however a considerable amount of pyrldine remained. Re-chromatography of the fraction did not remove the pink colouration and attempts to remove the remaining pyridine by prolonged freeze drying or by azeotroping the pyridine with a small amount of benzene or toluene were unsuccessful. Therefore a repeat extraction was performed to obtain sufficient compound for structural elucidation and all subsequent NMR spectra were obtained with Gfe-DMSO as the solvent
Fraction F70.2.1 was not fractionated further as H-NMR suggested that it consisted largely of compounds carried over from fraction F70.1. No further separation of fraction F70-2.2 was possible using C18 or C8 columns with MeOH or MeCN in either H2O or 1% TFA as the mobHe phase. However, preliminary 1H-NMR indicated the presence of aromatic signals. Separations using phenyl bonded silica are based on p interactions. Therefore a phenyl column (5 cm) was used to separate fraction F70.2-2 into eight fractions as shown in FIG 30 and Table 8. Although the fractions obtained from F70.2.2 were mixtures, peaks F70.2.2.3, F70.2.2.6 and F70.2.2.7 were predominantly single compounds by 1H-NMR. However insufficient material was available to re-chromatograph the fractions in order to obtain enough pure compound for structural elucidation.
Of the remaining peaks, only F70.2.5 was obtained in sufficient quantity to attempt further separation. Attempts to separate F70.2.5 using C18 and C8 columns were unsuccessful while phenyl bonded silica (5cm) resolved F70.2.5 into eight peaks using a MeOH in 1 % TFA gradient (FIG 31 and Table 9).
Total yield for this fraction was lower than expected from previous separations. However It was noticed that F70.2.5 did not completely dissolve in MeOH and therefore some material was lost during filtration. Of the eight peaks obtained from F70.2.5 only F70.2.5.2 was isolated in sufficient purity

and quantity to attempt a structural assignment. Peak F70.2.5.3 was also isolated in sufficient quantity, however the peak contained two compounds by 1H-NMR which could not be further resolved. Although by 1 H-NMR peaks F70.2.5.5 and F70.2.5.6 were predominantly single compounds, they were found to contain several compounds when further fradionation was attempted. Consequently Insufficient material was available for structural elucidation, Fraction 70.3.
As was the case with F70.2, fraction F70.3 could not be satisfactorily separated using MeOH/1 % TFA mixtures. Again MeCN in 1 % TFA gradients were used and F70.3 was separated into seven fractions (FIG 32 and Table 10) using a semi-preparative C18 silica (25 cm). Again the separation could be monitored successfully at 233nm.
Although no compounds were 100% pure by HPLC or1H-NMR at this stage, repeated chromatography using identical conditions aliowed the final isolation of five peaks (F70.3.2, F70.3.5, F70,3.6 and F70.3.7). FIG 4.18 shows chromatograms of fractions F70.3.5 and F70.3.7 indicating the peaks which contained pure compounds as shown by1 H-NMR immediately after drying the samples. It should also be noticed that even though the peak Isolated as F70.3.7 appeared to be pure by 1 H-NMR and by HPLC, the compound appeared to decompose, producing peaks which eluted at the same retention times as F70.3.6 and F70.3.5 (FIG 33).
Using MeCN/1 % TFA as the mobile phase on a C18 column (25 cm), F70.3.4 was separated into eight peaks as shown in FIG 34 and Table 11. In all eight peaks were collected, however only compounds F70.3.4.2 and F70.3.4.5 were obtained in sufficient quantity or purity to allow structural assignment.
The ^-NMR of fraction F70.3.3 showed that it consisted of three major compounds. This peak was subjected to further chromatography in order to separate these compounds however no further resolution could be obtained.
Fraction F70.4.

An initial separation of F70.4 was achieved using a 70%MeOH/1% TFA gradient on a C18 semi-preparative column (25 cm) (FIG 35 and TaWe 12). As can be seen traction F70.4.2 consisted of a single, large peak which upon closer examination, by both HPLC and 1H-NMR, was found to consist of several compounds. Further separation of F70.4.2 was not obtained with C18 or C8 columns using MeOH or MeCN gradients and as 1H-NMR showed the presence of aromatic protons in F70.4.2 a phenyi column (5 cm) was used. An isocratic mobile phase of 35%MeCN in 1 % TFA separated F7&4.2 into four major peaks (FIG 36 and Table 12A).
Repeated chromatography using identical conditions led to F70.4.2.2, F70.4.2.3 and F70.4.2.4 being obtained in sufficient quantity and purity to begin structural assignment It was obvious from the FIG 35 and also from 1H-NMR that F70A3 contained more than one compound. No improvement in separation of this fraction could be achieved using C18 or C8 reverse phase columns with MeOH or MeCN gradients, however a phenyi column (5 cm) again provided the separation. Although a separation could be obtained using either MeOH or MeCN gradients as the mobile phase, the separation obtained using MeCN in 1% TFA gradient provided better resolution of the peaks (FIG 37 and Table 13).
As can be seen from the chromatogram In FIG 37, F70.4.3.1 was, as expected, a mixture of compounds. At least five peaks were observed in the chromatogram, however further separation on this fraction was not attempted. Repeated chromatography using identical conditions of fractions F70.4.3.2, F70.4.3.4 and F70A3.5 enabled these fractions to be obtained in sufficient quantity for structure elucidation. Fraction F70.4.3.3 was a mixture of mainly two compounds, however these compounds could not be successfully separated. Fraction F80 Preliminary separation of F80 was achieved using a C18 reverse phase column (25cm), the chromatogram is shown in FIG 38 and the yields in Table 14. Although six peaks were collected initially, it was noticed that F80.2 and F80.3 were unstable as evidenced by the loss of these peaks in

the ehromatogram over time (FIG 38). Consequently ft was not possible to obtain structural information of these compounds.
The peak which eluted as F80.4 dearly contained more than one compound as evidenced by the ehromatogram and confirmed by 1H-NMR. No further improvement in the separation of this fraction was obtained using MeOH or MeCN gradients using a 25cm C18 column. However a change to the shorter C18 column (5cm) enabled a separation of the fraction into six peaks using a MeCN/1% TFA gradient (FIG 39 and Table 15). The peak which eluted at F80.4.5 was subjected to further chromatography under identical conditions to yield a single pure compound (F80.4.5.2).
Repeated chromatography of F80.5 using identical conditions (C18 25 cm) resulted In the purification of a single compound.
Although chromatographically F80.6 appeared to be a single compound, tH-NMR suggested that this fraction was a mixture of several compounds. The presence of aromatic signals in the 1HrNMR of F80.6 suggested that a phenyi bonded silica column (5 cm) would be useful. Both MeOH and MeCN In 1% TFA were employed as mobile phases and ft was found that the best separation was obtained with an isocratic gradient (40% MeCN in 1% TFA). This allowed the separation of F80.6 Into eight fractions (FIG 40 and Table 16). From these eight peaks it was possible to obtain five pure compounds (F80.6.2, F80.6.3, F80.6.4, F80.6.6 and F80.6.7) in sufficient quantify to attempt a structural assignment. Summary,
The current project isolated 21 major and minor peaks to either purity or to sufficient purity to be able structural elucidation of the compounds. Table 17 shows the compound numbers and weights isolated. Before structural elucidation was commenced each of the compounds were checked by HPLC, under identical conditions to those in which they were isolated, In order to confirm their state of purity.
As can be seen from the previous discussion, in addition to the compounds shown in Table 17, a large number of fractions were Isolated during the course of the current project Many of these fractions contained a

large number of compounds as evidenced by 1H-NMR and by the appearance of the chromatograms (i.e. a number of unresolved peaks). Several of these peaks appeared to contain only a small number of compounds, in some cases only one or two. as determined by 1H-NMFL However, insufficient plant material was made available to continue purification of these fractions. Table 18 presents some information on those tractions which would merit further investigation should more plant material become available tor future work.
The assignment of a structure as a saponin requires identification of each of the component parts of the molecule and the sequence in which they exist. These considerations can be expressed as follows :-
- the structure of the genuine agtycone
- the number of sugar residues
-the nature and sequence of the sugars in tie monosaccharide chain
- the anomeric configuration of each sugar unit
- the configuration and conformation of the irrtergtycosjdic linkage
- the attachment of the carbohydrate, chain to the aglycone
- the nature and position of any ack) in the molecule [1,2,6J.
However it should be noted that although it is possible to obtain complete
structures of saponlns using NMR alone, confirmation of the assignment
should, where possible, be provided by other methods.
Structural elucidation was carried out using known UV, IR, MS and NMR analysis. Fraction F70.3.6 - Chemical Elucidation
One of the first compounds Isolated In the current work was F70.3.6, which was obtained in relatively large yield (152.6mg). The following is a summary of the techniques and methods used to assign a structure to F70,3.6. Materials and methods.
The materials and methods used for the Isolation and structural

elucidation of F70.3.6 are presented in detail In Section B.
Hydrolysis was achieved by refluxing a 10mg sample In 1M HCI (Ajax). The solution was cooled and extracted with three 5mL volumes of dlchloromethane (DCM). The aqueous layer was neutralised wfth Ag2CO3 and filtered. Several mobile phases were tried to give a separation of the sugars and a chlorofoTm(CHCJ3):MeOH:aceticacid (AcOH):H2O (7:3:1:0.5) mixture proved successful on silica TLC plates (FIG 41). Standard sugar solutions used were ft-D-glucuronlc acid, (J-D-fucose, fi*D-glucose, V-L-arabinose, B-D-galactose, V-L-rhamnose, fl-D-ga!acturonic acid and fi-D-xyfose.
The TLC plates were developed with a phenol-sulphuric add solution (dissolve 3g phenol and 5mL 97% H2SO4 in 95 mL ethanot). The ptetea were dipped in the solution and heated at 110°C for until spots visualise (10-15 minutes). Instrumentation.
The methods and instrumentation used for NMR and ESMS are presented in detail in Section B. Compound F70.3.6 was also subjected to fast atom bombardment (FAB) MS in both positive and negative ion mode (Kratos Concept iSQ High Resolution/Quadrupote Tandem Mass Spectrometer, Central Science Laboratory University of Tasmania). The matrix used was meta-nitro benzyl alcohol (MNBA).
Sub-fractions from 70.2.5 were analysed and assigned structures through the use of spectra, in particular the 1H, 13C and ^^C-gHSQC (gHSQC), in a manner similar to that used above for traction 70,3.60. The resultant structures fell within three categories, aglycones, monodesrnosides and bidesmosides. The description of their structures begins with the aglycones. Where sugar residues in the mono- and bt- desmosWes had identical structures and relative stereochemistry to F70.3.6, the absolute stereochemistry was also assumed to be the same. UV spectrum.
Little structural information could be determined from the UV (Figure 42) or the FTIR (Figure 43) spectra. The UV spectrum shows two strong

absorbances at 205.9 nm (Abs. = 1.78, e« 11226) and 229.8 nm (Abs. = 2.28, & 14430) and two weaker absorbances at 274.6 nm (Abs.» 0.17, e& 1076) and 280.0 nm (Abs. = 0.14, e$ 886)(Figure 41). These absorbances are consistent with K bands (B-*B* transitions) and B bands in an aromatic
FTIR spectrum.
The FTIR of a thin film of neat compound te shown in FIG 43, The broad band centred at 3400 cm"1 indicates OH stretch and the group of peaks between 2820 and 3000 cm"1 suggest C-H stretch. Ester carbonyl stretch was shown by the strong peaks at 1723 and 1709 cm"1, white C-0 stretch was seen between 1275 and 1039 cnrf1. NMR methods.
Initially tt was decided to use NMR solvents which were reported in the
literature, principally methanoi (MeOH-d«) and pyridine-cfe. However it soon
became apparent that both solvents presented problems with the saponins
Isolated in the current work. The two problems encountered with MeOH-d*
were solubility and the loss of exchangeable protons. Several of the
saponins which were Isolated were not soluble in MeOH-cf* and of those that
were, several precipitated in the NMR tube over time. The presence of a
large number of overlapping signals in the 1H-NMR spectrum did not allow
for assignment of many signals, jn particular those which were due to the
carbohydrate moiety.
Signals due to exchangeable protons may contribute to or complicate the overall structural information. However in this instance it was decided that exchangeable protons may assist in the final structural assignment, although more information was added to an already complex spectrum. Therefore pyridfne-cfe, which seems to be the solvent of choice in the literature, was investigated. However the solvent signals tended to overlap the aromatic signals in the spectrum. Removal of the solvent also proved problematic. The solvent was removed in vacuo by freeze drying, however a substantial amount remained, even after several days. The solvent also appeared to react with some of the compounds to produce a bright pink

complex. It was not clear from the 1 H-NMR whether the compound Itself had undergone any changes but it could be seen that a significant amount of pyridlne remained. The addition of a small amount of benzene or toluene to azeotrope the pyridine, either by rotavap or freeze drying, did not resolve the problem.
Finally it was decided to use dimethyl sulfoxlde (DMSO^/e) as the NMR solvent, although little literature information was available for comparative purposes. The 1H-NMR spectrum of compound F70.3.6 in DMSO-de can be seen in FIG 44. The spectrum can be broadly divided Into three regions. The first most shielded region (*0.00 - *2.20) is predominantly associated with methyl signals. Six methyl singlets are evident in this region at chemical shifts of *0.71 (3H), *0.79 (3H), *0.84 (3H), *0.94 (3H), *0.97 (6H)and*1.33(3H).
The second region (*2.20 - *6.00) contains protons in close proximity to carbons bearing oxygen, notably four doublets characteristic of anomeric protons at *7.43, *7.52, *7.59, *7.66, *7.89 and *7.94. Finally the third, most deshielded region (*7.40 - *8.00), Indicates six aromatic protons at ♦7.43, *7.52, *7.59. *7.66, *7.89 and *7.94.
The 13C spectrum (FIG 46) was less crowded than the 1H spectrum and readily allowed identification of several carbon types. These deluded five carbonyl caibons (♦164.2, *164.3, +168.5, *169.6 and *171.9), two olefinic carbons (+121.6 and *141.7), eight aromatic carbons (*127.8, *127.9, *128.4. *128.5, *129.1, *129.5, *132.5 and *132.8) and four anornerlc carbons (*101.8, +102.4, *102.8 and V103.4).
Each region of the compound of F70.3.6 was analysed using 1H, 1H-. dqfCOSY (dqfCOSY) and 13C-gHMBC (gHMBC) spectral analysis.
The assignment of chemical structures, including stereochemistry, were based on the NMR and other spectra obtained were carried out using known techniques. The fraction of F70.3.6 was assigned the structure of FIG 46, shown below.

Mass Spectrometrv.
The structure of F70,3 Information In both positive and negative ion modes was obtained using electrospray (ES) as the lonisation source. Electrospray Is a soft ionfsation technique and use of ES ensured that minimal fragmentation of the molecule occurred under normal running conditions. The determination of an accurate mass was also possible using high resolution ES-MS. It was found that in positive ion mode the molecular ion was not as pronounced as in negative ion mode however more fragmentation was observed in positive ion mode. It Is not uncommon for compounds In ES to form adducts, in particular with sodium in positive ion and chlorine m negative Ion mode. This needed to be considered when assigning any peaks In the MS to fragments. The literature reports most MS In negative Ion mode for saponins and therefore this was the first mode investigated. The negative Ion high resolution (HR) ES-MS showed all peaks as pairs indicating the monoisotopic mass and the mass of the compound containing one 13C (FIG 47). The parent Ion presented as two peaks at m/z values of 1441.6472 and 1442.6529. This mass is consistent with a molecular formula of C73Hi02O29 (calculated 1442.6507) which supported the structure as assigned by NMR methods.
Other fractions where analysed in a similar manner to F70.3.6 above and the following compound structures were elucidated.

Compound F7Q.2.5.2: - This compound was isolated as 7.2 mg of an
amorphous white powder.
A compound refated to F70.2.5.2,2a,3p.19a-trihydroxy-olean-12-ene-23,28-dioic acid 28-O-p-D-glucopyranoside. was previously isolated from Bamngtonia acutangula [1]. This compound differs from F70.2.5.2 in that the acid at C2s has a glucopyranoside moiety. Therefore the present compound rs 2a 3p, 19a~trihydroxy-oiean-12-ene-24, 28-dioic acid (FIG 48).
A second aglycone (F70.2.5.3) was isolated in the current project as 1.4 mg of a white substance. The mass of the compound was m/z 485.2912 ([M - I]"), which is consistent with the molecular formula C3oH4605 (calculated m/z 486.3345). This suggests the loss of two hydroxyl groups from F70.2.5.2, however insufficient material could be obtained to provide further structural information.
Monodesmosktes - The monodesmosidic compounds described in the following section are grouped according to the functionalities present at C2i
and C22 of the aglycone.
Benzoate at C?i and hvdroxyl at C22.
Compounds F70.2.3.2 and F70.3.2 were shown to have a benzoate moiety at C21 and a hydroxyl group at C22- Both were isolated as amorphous white solids in low yield (8.5 and 26-3 mg respectively).
Benzoate at CM and feo-butyrate G??.
Compound F70.3.4.2 - This compound was isolated as 117 mg of an
amorphous white powder.
Benzoate at both C21 and C22: Compounds in this group each had a
benzoate functionality at both C21 and C22- Four such compounds were
isolated in the current project. Each of the four compounds were isolated
as amorphous white substances and the weights were F7QA3.5.2 (2.7
mg), F70. (4.9 mg), F80.6.4 (11.7 mg) and F80.6.7 (3.9 mg).
Benzoate at C21 and tiglate at C22. - The compounds in this group were
characterised by a benzoate at C21 and a tiglate at C22 Amended Sheet 1PEA/AU

compounds were isolated in the current project, each as an amorphous white mass, F70.4.2.3 - 42.7mg; F70. - 6.4mg; F80.6.3 - 25.8mg; F80.6.6 - 4.3mg.
Tiglate at both C21 and C22 - The two compounds isolated in this series, F70. (2.8mg) and F80.6.2 (7.4mg) were characterised by tiglate groups at both C21 and C22.
Additional compounds:
Fraction F70.3.3 was isolated as 47.8mg of an amorphous white mass which was shown to contain several compounds by 1D and 2D NMR. Several attempts to resolve these compounds resulted in the collection of 6.6mg of fraction F70., which still contained three compounds by 1D and 2D NMR. However it was possible to assign the structure of one of these compounds using a similar approach as for the other compounds.
Bidesmosides - Several compounds which had a sugar moiety at C?t were
isolated in the current project, in all instances this sugar was assigned as an
Compound F70.2.6.2: In comparison to the other compounds isolated in
the current project a relativery high yield was obtained for F70.2.6.2. This
compound was isolated as 38.5mg of an amorphous white powder.
Compound F70.3.4.5; A small quantity (2.7 mg) of F70.3.4.5 was isolated
as an amorphous white compound. The structure of F70.3.4.5 is shown In
FIG 60.
Compound F70.3.5.2:
This compound was isolated as 12.2 mg of an amorphous white compound.
Compound F70.3.7.2:
Compounds F80.4.5.2 andF80,5.2:
These compounds were isolated as 2.8 mg (F80.4.5.2) and 20.1 mg (F80.5.2) of amorphous white masses.

This section summarises the structural assignment of the compounds isolated in the current project. In all one agiycone, ton monodesmosides and six bidesmosides were isolated and characterised. A search of the available literature based on either structure or molecular weight felled to find any of the mono- or bi- desmosides and they are at the time of writing assumed to be novel structures. The aglycone is, as mentioned a known structure from Banington/a acutangula. There were many minor compounds in the extracts which also need to be characterised (see previous section). This would be facilitated by the collection of a substantial amount of bark and large scale preparative chromatography.
SECTION D - Analgesic activity of F70.3.2 and F70.3.6
Preliminary investigation into whether intravenous administration of
the test compounds F70.3.2 and F70.3.6, produce pain-relieving effects in a
rat model of inflammatory pain.
AduR mate Sprague-Dawley rats were housed in a temperature controlled room (21 ± 2° C) with a 12h/12h light/dark cycle and free access to both food and water. Ethical approval for this study was obtained from the Animal Experimentation Ethics Committee of The University of Queensland. Reagents and materials
Isoflurane (Forthane®) was obtained from Abbott Australasia Pty Ltd
(Sydney, Australia). Sodium benzylpenicillln vials (600 rrig) were purchased
from CSL Ltd (Melbourne, Australia). Normal saline ampoules were obtained
from Delta West Pty Ltd (Perth, Australia) and heparinised saline (50 IU/S
ml) was purchased from Astra Pharmaceuticals Pty Ltd (Sydney, Australia).
Single lumen polyethylene tubing {I.D. 0.5 mm, O.D. 1.00 mm) was
purchased from Auburn Plastics and Engineering Pty Ltd (Sydney, Australia),
Sterile slficonized silk sutures (Dysllk™) were obtained from Dynok Pty Ltd
(Adelaide, South Australia).

induction of hindpaw inflammation
Inflammation of the rat hrndpaw was induced by fntra-plantar (i.pl.) injection of Freund's Complete Adjuvant (FCA, 0.15 mL) whilst rats were under brief 3% isoflurane: 97% oxygen jnhalational anaesthesia. Inflammatory pain was assessed using the paw pressure test (see below for details). The test compounds were administered on either day 5 or day 6 following I.pl. FCA administration. Paw Volume Measurement:
The volume oi each of the left and right hindpaws was measured using a plethysmometer on the day prior to Lpi. FCA administration, as well as immediately prior to J.v. drug (or saline) injection on day 5 post-FCA administration. Surgery Jugular Vein Cannulation
Jugular vein cannulation was performed whilst rats were anaesthetised with 3% isoflurane: 97% O2 inhalational anaesthesia maintained using a calibrated Trilene vapouriser. Polyethylene cannulae (previously filled with heparinised saline) were implanted into the right common jugular vein for i.v. drug dosing. Cannulae were exteriorized by subcutaneous (s.c.) tunneling to an incision made in the interscapular area, and protected by a stainless steel spring, the base of which was positioned in a subcutaneous pocket between the scapulae. Incisions were closed with sterile silk sutures. After surgery, rats were housed singly in metabolic cages and were allowed to recover post-operatively for a minimum of 2 h before further experimentation. Food and water were available ad libitum during the recovery period. Drugs Administered
Each of the compounds F70.3.2 and F70.3.6 were dissolved in sterile saline. The compounds of interest were administered initially in a dose of 0.01 mg/kg in an injection volume of 0.5 mL, The other Lv. doses administered in this preliminary study were as follows:
F70.3.2 0.002 mg/kg, 0.005 mg/kg, 0.01 and 0.02 mg/kg

F70.3.6 0.002 mg/kg, 0.005 mg/kg, 0.01 mg/kg. 0.02 mg/kg and
0.05 mg/kg
Control rats received bolus doses of i.v. saline. Antinociceptive Testing: Paw Pressure Test
For the paw pressure test, rats were gently restrained under a towel, and incremental mechanical pressure (maximum=250 g) was appied to the dorsal surface of the hindpaw. The pressure required to elicit paw withdrawal, the paw pressure threshold (PPT), was determined. The mean of 3 consecutive measurements, separated by 10 s, was determined. The same procedure was then performed on the contralatera! side with the sequence of sides being alternated to preclude order effects. Similarly, rats were gently restrained under a towel for the quantification of paw volumes using a ptethysmometer. Data Analysis
Following administration of i.v. bolus doses of each of the test compounds (F70.3.2 and F70.3.6), paw withdrawal thresholds (g) were normalized by subtraction of the individual baseline PWT values quantified immediately prior to drug administration. The area under the normalized PWT versus time (AUC) was calculated using the trapezoidal rule. Dose-response curves were constructed by plotting the AUC values versus the i.v. dose for each of F70.3.2 and F70.3.6-
FIG 65 shows that for the ipsiiateral hindpaw, there was a dose-related increase in the antinociceptive potency of F70.3.2, By contrast there was an absence of antinoctception in the contralateral hindpaw.
FIG. 66 shows that the ipsiiateral hindpaw, there was a dose-related Increase in the antinociceptive potency of F70.3.6 in the dose range, 0.002-0.01 mg/kg. However, further increases in the dose magnitude resulted in a decreased rather than an increased antinociceptive response. Similar to compound F70.3.2, there was an absence of antinociception in the contralateral hindpaw.
The results summarised in FIG 67 In which rats where injected with saline onfy show that the experimental procedures themselves did not evoke

significant arrtinociceptfon
The is a distinct increase and then plateau of the antinocleptive effect of compounds F70.3.6 and F70.3.2 using the above techniques, as shown in FIG. 68.
FIG 69 shows that by 5 days post-FCA administration, the mean (± SEM) volume of the ipsilatera) hlndpaw increased approximately 2-fold from 3.3 (± 0.1) mL to 6.0 (± 0.1) mL By contrast, the mean (± SEM) volume of the contrafateral hindpaw at day 5 post-FCA administration (3.3 ± 0.1 mL) did not differ significantly from that ((3.2) ±0.1 mL) measured prior to induction of Inflammation in the ipsliateral hindpaw.
A range of side effect were observed forth© compounds F70.3.2 and F70.3.6 as summarized below In TABLE 19 and 20.
Throughout the specification the aim has been to describe the preferred embodiments of the invention without limiting the invention to any one embodiment or specific collection of features.
Throughout this specification, unless the context requires otherwise, the word "comprises", and variations such as 'comprise' or "comprising", will be understood to Imply the inclusion of a stated interger or group of integers or steps but not to the exclusion of any other interger or group of integers. Reference may also be made to FIGS 70 and 71 which show the structures of further compounds of the invention. FIG 70 shows compounds (1) to (7) characterized by inclusion of groups A, B, C and D as shown. These compounds have been prepared from water extracts of the dried bark of B acutanqula as described previously and have had their structure assigned as also described previously. Similar comments apply to the compounds of FIG 71.
1. Payens, J.P.D.W, (1967). A monograph of the genus Baningtonia
(Lecythjdaceae). Blumea 15(2): 157-263.
2. Everist, E.L. (1974) Poisonous plants ofAustnatia. Angus & Robertson
3. Quisumbing, B. (1978) Medicinal plants of the Philippines. Quezon

City. Katha Publishing.
4. Carr, T. (1947). Barringfonia acutangula as fish poison. A practical
application. North Queensland Naturalist 15:3-4,
5. Cox, P A (1979). Use of indigenous plants as fish poisons in Samoa.
Economic Botany 33(4): 397-399.
6. Lahiri, J.K. and Ghosh, S. (1942). Chemical examination of the seeds
of Barringtonia acutangula Gaertn. Journal of the American
Pharmaceutical Association 31(7): 193-194.
7. Webb, LJ. (1959). The use of plant medicines and poisons by
Australian Aborigines. Mankind 7:137-146.
8. Barua, A.K., Dutta, S.P. and Pal, S.K. (1967). Chemical examination
of Barringtonia acutangula Gaertn.
9. Sharma, B.M. and Singh, P. (1976). Pharmacognostic study of fruits
of Barringtonia acutangula Gaertn. Herba Hungarica 15(3): 7-14.
10. Cribb, A.B. and Cribb, J. W, (1982). Useful wild plants In Australia..
Sydney. Fontana Books.
11. Webb, LJ. (1948). Guide to the medicinal and poisonous plants of
Queensland. Council for Scientific and Industrial Research Bulletin
no. 232.
12. Watt, J.M. and Breyer-Brandwijk, M.G. (1962). The medicinal and
poisonous plants of Southern and Eastern Africa. Being an account
of their medicinal and other uses, chemical composition,
pharmacological effects and toxicology in man and animal. 2nd ed..
Edinburgh and London. E and S Livingstone Ltd.
13. Uphof, J.C.T. (1968). Dictionary of economic plants. 2nd edition ed..
Wtlrzburg. Richard Mayr.
14. Jebb, M. (1992). Edible Barringtonlas. Kew Magazine 9(4): 164-180.
15. Tanaka, T. (1976)- Tanaka 's cyclopedia of edible plants of the world.
ed. S. Nakao. Tokyo. Kefgaku Publishing Company.
16. Cribb, A.B. and Cribb, J.W. (1981). Wild medicine In Australia. .
Sydney. Fontana Books.
17. Worseley, R.R.L. (1934). The Insecticldal properties of some East

African plants. Annals of Applied Biology 2±: 649-669.
18. Lassak, E.V. and McCarthy, T. (1990). Australian medicinal plants..
Port Melbourne. Mandarin Australia.
19. Ahmad, KJ. (1969). Pharmacognosy of the leaf and root of
.Baningtonia acutangula, Gaertn. Planta Medica 17(4): 338-345.
20. Kind, F.A. and Gedeon, J. (1957). About saponins and sapogenins
of Barringtonia species. Archiv der Pnarmazte Berlin 289(61 V. 140-
21. Nozoe, T. (1934). Polyterpenoids and their glycosides. I. Saponins
from the seeds of Barringtonia asiatica Kurtz. Journal of the Chemical
Society of Japan 55: 1106-1114.
22. Nozoe, T. (1934). Polyterpenoids and their glycosldes. II.
Constituents of the sugar part of A Society of Japan 55:1115-1123.
23. Nozoe, T. (1935). Polyterpenoids and their glycosides. III. A-j-and
A2-barrigenoL Journal of the Chemical SocietyofJapan 56:689-703.
24. Cole, A.R.H., Downing. D.T., Watkins, J.C. and White, D.E. (1955).
The constitution of A-j -Barrigenol. Chemistry and Industry; 254-255.
25. Anantaraman, R. and Pillai, K.S.M. (1956). Barringtogenol and
oarringtogenic acid, two new triterpenold sapogenins. Journal of the
Chemical Society: 4369-4373.
26. Barua, A.K, Marti, P.C. and Cnakraborti, S.K (1961). Triterpenoids
XI. New triterpenold sapogenins from the fruits of Barringtonia
acutangula. Journal of Pharmaceutical Sciences 50(11): 937-940.
27. Barua, A.K., Dutta, S.P. and Das, B.C. (1968). Triterpenoids - XXIX.
The structure of Barringtogenof B - A new triterpenoid sapogenin from
Baningtonia acutangula Gaertn. Tetrahedron 24:1113-1117.
28. Barua, A.K,, Dutta, S.P. and Pal, S.K. (1967). Triterpenoids - XXX.
The structure of Barringtogenol E - A newtriterpenoid sapogenol from
Barringtonia acutangula Gaertn. Journal of the Indian Chemical
Society 44{U): 991-993.

29. Chakraborti, S.K. and Barua, A.K. (1962). Triterpenoid XIIJ. The
constitution of Barringtogenol D. Experientfa 18_:66-67.
30. Chakraborti, S.K. and Barua, A.K. (1963). Triterpenes - XVI. The
constitution of Barringtogenol D - A new triterpenoid sapogenin from
Barringtonia acutangula Gaertn. Tetrahedron 19:1727-1732.
31. Barua, A.K. and Chakrabarti, P. (1965). Triterpenoids - XFX The
constitution of Barringtogenol C - A new triterpenoid sapogenin from
Barringtonfa acutangula Gaertn. Tetrahedron 21: 381-387.
32. Barua, A.K. and Chakrabarti, P. (1964). Triterpenoids XVIH. The
constitution of Barringtogenol C. Science and Culture 30(7): 332-334.
33. Barua, A.K., Chakraborti, S.K., Chakrabarti, P. and Maiti, P.C. (1963).
Triterpenoids. Part XIV, Studies on the constitution of
Barringtogenol C * A new triterpenoid sapogenin from Barringtonia
acutangula Gaertn. Journal of the Indian Chemical Society 40(6):
34. Chakrabarti, P., Pal. S.K, and Barua, A.K. (1967). Terpenoids. Some
reactions of Barringtogenoi C. Proceedings of the Indian Science
Congress 54^; 149.
35. Barua, A.K., Basak, A. and Chakravarti, S. (1976). Triterpenoids.
XLIV. The revised structure of Barringtogenol B. Journal of the
Indian Chemical Society LJH: 209-210.
36. Barua, A.K., Chakrabarfi, PM Gupta. A.S.D., Pal, S.K., Basak, A.,
Banerjee, S.K. and Basu, K. (1976). The structure and
stereochemistry of barrlgenic acid, a new triterpene acid from
Barringtonia acutangula. Phytochemistry 15:1780-1781.
37. Barua, A.K., Pal, S.K. and Dutta, S.P. (1968). Triterpenoids - XXXI.
Studies on a triterpene isolated from Barringtonia acutangula Gaertn. Science and Culture 34(6V. 259-260.
38. Narayan, G.K.A.S.S., Row, LR. and Sastry, C.S. (1976). Chemical
examination of leaves of Barringtonia acutangula Gaertn. Current
Science 45(14): 518-519.

39. Dhaveji, K., Narayan, G.K.A.S.S., Rao, D.S. and Row, LR. (1984).
13C NMR spectra of Acutanguiic and Tangulfc acids from
Barringtonia acutangula Gaertn. Journal of the Indian Chemical
Society IM 1032-1033.
40. Anjaneyulu, A.3.R.. Sastry, C.S.P., Narayan, G.K.A.S.S. and Row,
LR. (1978). New Iriterpenes from Barringtonia acutangula Gaertn.
Journal of the Indian Chemical Society LV(11): 1169-1174.
41. Pal, B,C>, B4 A. and Price, K.R. (1991). A triterpenoid glucoside from
Barringtonia acutanguia, Phytochemistry 30(12): 4177-4179.
42. Pal, B.C-, Chaudhuri, T., Yoshikawa, K. and Arihara, S. (1994).
Saponins from Barringtonia acutangula. Phytochemisfry 35(5): 1315-
43. Gupta, M.B., Bhalla, T.N., Gupta, G.P., Mitra. C.R. and Bhargava,
KP. (1969). Anti-inflarnmatory activfty of natural products (I)
Triterpenoids. European Journal of Pharmacology &. 67-70.
44. Aquino, R., De Feo, V., De Simone, F., Pizza, C. and Cirino, G,
(1991). Plant metabolites. New compounds from and anti-
infiammatory acti\1ty of Uncaria tontentosa. Journal of Natural
Products 54(2): 453-459.
45. Scalbert. A. (1991). Antimicrobial properties of tannins.
Phytochemistsv 30(12); 3875-3883.
46. Hiller, K, (1987). New results on the structure and biological activity
of triterpenoid saponins, in Biologically active natural products. K.
Hostettmann and P.J. Lea Editors. Clarendon Press. Oxford.
47. Hostettmann, K. and Marston, A. (1995). Saponins. Chemistry and
pharmacology of natural products, ed. J.D. Phiilipson, D.C. Ayres,
and H. Baxter. Cambridge. Cambridge University Press.
48. Chen, S. and Snyder, J.K. (1993), General strategy for the
determination of'saponins: MoUusdcktal saponins from AlHum vineale,
in Bioactive natural products. Detection, isolation and structural
determination. S.M. Colegate and R.J. Molyneux Editors. CRC
Press. Boca Raton.

49. Massiot, G. and Lavaud, C. (1995). Structural elucidation of
saponjns. Studies In Natural Product Chemistry 15:187-224.
50. Mailtard, M., Marston, A. and Hostettmann, K. (1993). Search for
moKusdddal and iarmldal agents from plants, in Human medicinal
agents from plants. A.D. Kinghom and M.F. Balandrin Editors.
American Chemical Society. Washington DC.
51. van Middtesworth, F. and Cannell, R.J.P. (1998). Complication and
partial identification of natural products, m Natural products isolation.
R.J.P. Cannell Editors. Humana Press. Totowa.
52. Cordell, G.A., Lyon, R.L., Fong, H.S., Benoit, P.S. and Farnsworth,
N.R. (1977). Biological and phytocbemical investigations ofDtenthu$
barbatus cv. "China Doll" (Caryophyfiaceae). Uoydla 40(4): 361^363.
53. Huang, H., Huang, N. and LI, S. (1982). Analgesic effect of saponin
from Dolichos falcatus. Yaoxue Tongbao 17(2): 122 (Ch).
54. Racz-Kotilla, E-, Petre, M. and Racz. G. (1982). Anti-noclceptive
effect of Platycodon grandifforum extracts. Reviews of Medicine
28(2): 180-182.
55. Lei, W., Shi, Q. and Yu, S, (1984). Analgesic and CNS inhibitory
effects of total saponins from the leaves of Panax notoginseng.
Zhongyao Tongbao 9(31:134-137.
56. Oshima, Y., Ohsawa, T.. Oikawa, K, Konno, C, and Hikino, H. (1984).
Structures of dianosides A and B, analgesic principles of Dianthus
superbus var. hng/calycinus herbs. Planta Medica 50f 40-43).
57. Oshima, Y., Ohsawa, T. and Hikino, H. (1984). Structure of
dianosides C, D, E and F, triterpenoid saponins of Dianthus superbus
var. longkalydnus herb. Planta Medica 50:43-47.
58. Oshima, Y., Ohsawa, T. and Hikino, H. (1984). Structures of
dianosides G, H and I, triterpenoid saponins of Dianthus superbus
var, longicalycinus herbs. Planta Medica 50:254-258.
59. Gomes, A., Sharma, R.M. and Ghatak, B.J.R. (1987).
Pharmacological investigation of a grycosldal fraction isolated from
Maesa chisia D. Don var augustifolla Hook F and Th. Indian Journal

of Experimental Biology 25: 826-831.
60. Dubuisson, D. and Dennis, S.G. (1977). The formalin test: A
quantitative study of the analgesic effects of morphine, meperidine,
and brain stem stimulation In rats and cats. Pain 4:161-174.
61. Hunskaar, S. and Hole, K. (1987). The formalin test in mice:
dissociation between inflammatory and non-inflammatory pain. Pain
62. Hunskaar, S., Fasmer, OB. and Hole, K. (1985). Formalin test in
mice, a useful technique for evaluating mrld analgesics. Journal of
Neurosdence Methods 14:69-76.Cohen, H. (1944).
63. Murray, C.W., Porreca, F. and Cowan, A. (1988). Methodological
refinements to the mouse paw formalin test An animal model of tonic
pain. Journal of Pharmacological Methods 20: 175-186.
64. Shibata, M., Ohkubo, T., Takahashi, H. and Inoki, R. (1989). Modified
formalin test characteristic biphasic pain response. Pain 38: 347-
65. Covino, B.G., Dubner, R., Kosterlitz, H.W., Ltebseklnd, J.C.,
Sternbach, R.A., Vyklicky, L, Yamamura, H. and Zlmmermann, M.
(1980). Ethical standards for the investigation of experimental pain in
animals. Pain 9:141 -143.
66. Zimmermann, M. (1983). Ethical guidelines for investigations of
experimental pain Jn conscious animals. Pain 16:109-110.
67. Hostettmann, K. and Marston, A. (1995). Saponins. Chemistry and
pharmacology of natuml products, ed. J.D. Phillipson, D.C Ayres,
and H. Baxter. Cambridge. Cambridge University Press.
68. Cannell, RJ.P.. ed. Natural products isolation. Methods in
biotechnology, ed. J.M.Walker.Vol. 4.1998, Humana Press: Totowa.
69. Chen, S. and Snyder, J.K. (1993), General strategy for the
determination of saponins: Molfusciddalsaponins from Allium vineale,
in Bioactive natural products. Detection, isolation and structural
determination. S.M. Cotegate and R.J. Molyneux Editors. CRC
Press. Boca Raton,
R2 is selected from hydrogen, hydroxyl, O-alkyl, O-a!kenyl, O-benzoyl, G-
alkanoyL O-alkenoyl, O-aryl, O-heterocyclic, O-heteroaryl or

wherein R5 and R7 are independently be selected from hydrogen, alkanoyl,
alkenoyl, benzoyl or benzoyl alkyl substituted alkanoyl;
R3 is selected from hydroxyl, O-alkanoyl, O-alkenoyl, O-benzoyl, O-alkyl, 0-
alkenyf, O-aryl, O-heterocycJic or O-heteroaryl;
R4 is selected from -CH20H, COOH, CH2OCOCH3. COO alkyi, COO aryl,
CH2COO alkyl. COO-heterocyclict COO-heteroaryl. CH2-0 aryl. CH2O
heterocyclic or CH20 heteroaryl;
Re is selected from hydrogen or

Ri is selected from hydrogen or alkyl; or
pharmaceutically acceptable salts thereof, with the provisos that when:
R2 is OH, R3 is OH, R4 is CH2OH, and Re is xylopyranosyl, R, is not H;
R4 is CH2OH and R3 is O-alkanoyl R2 is not O-acetyl;.
R R4 is CH2OH and R3 is hydroxyl then R2 is not hydroxyl..
2. A compound as claimed in claim 1 wherein R2 is hydrogen, O-
benzoyl, O-tigloyl or

wherein R5 and R7 are selected from hydrogen, tigloyl, benzoyi or benzoyl alkyl substituted alkanoyl.
3. A compound as claimed in claim 1 wherein R3 fs selected from O-
acetyl, O-benzcy!, O-isobutyryl or O-tigloyl.
4. A compound as claimed in claim 1 wherein R4 is selected from
-CH2OH, O-acetyl or hydroxyl.
5. A compound as claimed in claim 1 wherein Rz is arabino pyranosyl, 3-
{3-benzoyI-2-methylbutanoyl)-4-benzoyl-a-L-arabinopyranosyl, 0-benzoyi,0-
dibenzoyl, O4ig[oyl, 3,4 dibenzoyl oc-L-arabinopyranosyl, 3-{3-benzoyl-2
methyfbutyryM-tigloyl-a-L-arabinopyranosyl. 3-tigloyI-4-(3-benzoyl-2
methylbutyryO-a-L-arabinopyranosyl, 3-(3»benzoyl-2 methyl butanoyI-4-
benzoyl-a-L-arabinopyranosyl or 3-(3-benzoyl-2 methylbutyryl)-4-benzoyt-a -
6. A compound as claimed in claim 1 wherein R6 is methyl.
7. 3-O-p-D-xylopyranosy!{1 ->3)-[P-D-galactopyranosyl(1->.2)]-(3-D-

arabinopyranosyl]-22-0-acetyl barringtogenol C;
8. 3-0-β-D-xyiopyranosyl(1->3HP-D-ga!actopyranosyl(1->2)]-P"D-
glucuronopyranosyl-21-O-benzoyl barringtogenol C;
9. 3-0-β3-D-xylopyranosyl(1->3Hp-D-galactopyranosyl(1^2)]-p-D-
glucuronopyranosyl'21-0-benzoy!-28-0-acetyl barringtogenol C;

10. 3-O-β-D-xylapyranosy!(1 ->3)-[P-D-galactopyranosyl(1 ->2)]-p-D-
gIucuronopyranosyI~21-0-ben20yl-22-0-isobutyry( barringtogenol C;
11. 3-0-β~D-xylopyranosyI( 1 -> 3Hp-D-galactopyranosyl(1 ->2)]-p-D-
methylglucuronopyranosyl-21T22-0-dibenzoyi barringtogenol C;
12. 3-0-βp-D-xylopyranosyi(1^3HP-D-galactopyranosyl(1->2)]-P-D-
glucuronopyranosyl-21, 22-O-dibenzoyl barringtogenol C;
13. 3-0-β-D-xylopyranosyl(1->3Hp-D-galactopyranosyl(1->2)]-3-D-
methylg!ucuronopyranosy!-21-O-ben2Oyl-22-O-tigloyl barringtogenol C;
14. 3-O-β-D-xylopyranosyI(1">3)-[(3-D-galactopyranosyl(1->2)- β-D-
glucuronopyranosyi-21-O-ben2oyl-22-O-tigloyl barringtogenol C;
15. 3-0-β-D-xylopyranosyl(1 ->3)-[p-D-galactopyranosyl(1 ->2)1-β-0-
methylglucuronopyranosyl-21f22-0-tigloyl barringtogenot C;
16. 3-0-β-D-xyIopyranosyl3)-[p-D-galactopyranosyl(1 ->2)]-D-
glucuronopyranosyl-21l22-0-tigloyl barringtogenol C;
17. 3-O-β-D-xyiopyranosyl(1 ->3)-[P-D-galactopyranosyl(1 ->2)]-£-D-
glucuronopyranosyl-22-O-benzoyl barringtogenol C;
18. 3-Q-β-D-xylopyranosyl(1 ->3)-[pD-gaIactopyranosyl(1 ->2)]-p-D-
barringtogenol C;
19. 3-O-P-D-xylopyranosyl(1 ^3)-[p-D-galactopyranosyl(1 ->2)]-p-D-
barringtogenol C;
20. 3-0-p-D-xy]opyranosyl(1->3)-(P-D-galactopyranosyi(1^2)]-p-D-
arabinopyrar»osyf]-22-O-acetyl barringtogenol C;
21. 3-0-P-D-xylopyranosyl(1 -»3)-[P-D-galactopyranosy[(1 ->2)]-P-D-

arabinopyranosyi]-22-O-acetyl barringtogenol C;
22. 3-0-βD-ga)actopyranosyl(1 ->2)-βOglucuronopyranosyl-21 -O-[3-(3-
barringtogenol C
23. 3-0-β-D-xy!opyranosyl(1->3Hp-D^a!actopyranosy!(1->2)]-P-D-
arabinopyranosyJJ-28-O-acetyl barringtogenol C.
24. A pharmaceutical composition for treatment and/or control of pain
comprising a therapeutically effective amount of a compound of any one of claims 1 to 23 and a pharmaceutically acceptable carrier.
25. A pharmaceutical composition as claimed in claim 24 wherein the
carrier is a pharmaceutically acceptable excipient.
26. A method of treating and/or controlling pain which includes the step
of administering to a subject in need of such treatment at least one
compound as claimed in any one of claims 1 to 23.



1851-CHENP-2006 AMENDED CLAIMS 07-02-2012.pdf

1851-CHENP-2006 FORM.3 07-02-2012.pdf

1851-CHENP-2006 OTHER PATENT DOCUMENT 07-02-2012.pdf

1851-CHENP-2006 POWER OF ATTORNEY 21-03-2012.pdf

1851-CHENP-2006 AMENDED CLAIMS 21-03-2012.pdf

1851-CHENP-2006 CORRESPONDENCE OTHERS 24-02-2012.pdf

1851-CHENP-2006 CORRESPONDENCE OTHERS 21-03-2012.pdf


1851-CHENP-2006 FORM-13 23-11-2006.pdf






1851-chenp-2006-form 1.pdf

1851-chenp-2006-form 13.pdf

1851-chenp-2006-form 26.pdf

1851-chenp-2006-form 3.pdf

1851-chenp-2006-form 5.pdf


Patent Number 252990
Indian Patent Application Number 1851/CHENP/2006
PG Journal Number 24/2012
Publication Date 15-Jun-2012
Grant Date 13-Jun-2012
Date of Filing 26-May-2006
Applicant Address PO Box 381, DERBY, Western Australia 6728
# Inventor's Name Inventor's Address
1 QUINN, Ronald Griffith University, Kessels Road, NATHAN, Queensland 4111
2 MILLS, Clive 44 Old Kilmainham Village, Bow Lane West, KILMAINHAM, DUBLIN 8
PCT International Classification Number C07H15/256,A61K31/351
PCT International Application Number PCT/AU2004/001660
PCT International Filing date 2004-11-26
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 2003906558 2003-11-27 Australia